3.2.1 First-Order Pictorial Elements of Visual Art

Several first-order pictorial elements of visual art that have received considerable investigation for many decades are the angularity (curved versus angular) and orientation (vertical versus horizontal versus diagonal) of lines and color. Nevertheless, there is still much to be learned about the contribution of these factors to an aesthetic experience with the use of research perspectives that improve upon flawed designs and stimulus materials of past studies. Representative investigations of the influence of fundamental or first-order pictorial elements on aesthetic judgments included the following. Silvia and Barona (2009) conducted a study to determine if people prefer curved over angular lines, shapes, and objects. They employed stimuli consisting of black-on-white displays composed of either seven circles or seven hexagons that varied in size and distribution within a display field creating different levels of structural imbalance of the composition’s pictorial elements. University students who differed in degree of self-reported training in the visual arts saw each stimulus for two  seconds and then rated its pleasingness. The researchers found that, overall, subjects preferred the displays composed of round circles more than angular hexagons and this was the case regardless of the level of imbalance of the arrays. However, expertise moderated the effect of angularity in that angularity reduced preference only for participants with lower levels of expertise. As expertise increased, the negative effect of angularity on pleasingness decreased. (The contribution of art expertise to an aesthetic experience is described in detail in Section 3.4.)

Not only has the angularity of lines within a composition been shown to influence preference for art, but line orientation also plays a significant role in preference. For example, Latto et al. (2000) demonstrated the existence of an ‘aesthetic oblique effect’ that is manifest by the preference of viewers for horizontal and vertical lines over oblique lines in visual art. They used eight of the well-known grid-based abstract paintings by the artist Piet Mondrian as stimuli. Six paintings are composed of high-contrast horizontal and vertical black lines extending either to another black line or to the edge of the pictorial field, and a red, yellow, and blue solid color area all on an off-white background; the other two paintings consist of bands of color. University students untrained in the visual arts viewed reproductions of the eight paintings shown separately for 5 seconds. Additionally each original was shown rotated seven times at 45° intervals resulting in half of the paintings being seen with horizontal and vertical line orientations and the other half with oblique component lines. Latto et al. (2000) found that paintings presented with their component lines in a horizontal or vertical orientation were rated as more aesthetically pleasing than when their lines were in an oblique position. This effect was enhanced when the lines were parallel to a picture’s format orientation (e.g. when vertical lines were in a vertically oriented painting). Additionally, the originals received slightly higher pleasingness ratings than their rotated versions. Latto et al.’s (2000) experiment and findings have since been replicated by Plumhoff and Schrillo (2009).

Latto and Russell-Duff (2002) report empirical evidence that artists are aware intuitively of the aesthetic oblique effect and use line orientation preferentially in their compositions. They measured the proportion of horizontal, vertical, and oblique lines in a representative selection of 88 twentieth-century paintings in the Israel Museum in Jerusalem that varied in painting styles, including abstracts, landscapes, and self portraits. The artists studied were found to overwhelmingly use vertical and horizontal lines as contrasted with oblique lines (45%, 31%, and 24%, respectively) to construct their compositions.

As mentioned, color is another perceptual fundamental of art that, despite a long history of empirically based investigations, continues to receive considerable attention by researchers interested in aesthetic beauty. As one example, Pozella et al. (2005) had university students untrained in the visual arts rate traditional and modern landscapes and portraits on 12 semantic differential scales (e.g. simple–complex, ugly–beautiful, no pleasure–extreme pleasure). Half the subjects viewed the paintings in color, the other half in black and white. The researchers found that removing color from portraits increased their perceived pleasantness and beauty, whereas removing color from landscapes reduced their perceived beauty. Thus, they concluded that the effect of color on viewers’ reactions to paintings depends on a painting’s content. Pozella et al. (2005) suggest the following explanation of this finding. With respect to portraits, recent findings concerning the neurophysiological mechanisms underlying face perception show that color and face perception reflect separate neuropsychological processes. This causes the presence of color to attenuate reactions to portraits because it provides superfluous information that adds to the processing load, conditions that may lessen a portrait’s aesthetic appeal. On the other hand, the color in landscapes may provide critical channels for transmitting increased amounts and types of information such as cues for depth, which facilitates the extraction, identification, and processing of critical information required for the perception of landscapes, thereby increasing aesthetic appeal.

3.2.2 The Organizational Structure of Artistic Compositions: Visual Rightness and Pictorial Balance

It is a widely held view among artists and writers on Western art that the meaning of a painting reveals itself only if a viewer sees it as an organized arrangement of pictorial elements – as a comprehensive structure (see, e.g. Arnheim, 1974). If a composition is to have maximum impact, it must be ‘visually right’ (‘good’); that is, there must be a certain order and equilibrium or balance among pictorial elements that is both visually right and expressive. The notion of visual rightness is implicitly present in many treatises on art. For example, Arnheim (1974) asserts that ‘for any special relationship between objects there is a “correct” distance, established by the eye intuitively. Artists are sensitive to this requirement when they arrange the pictorial objects in a painting or the elements of a piece of sculpture’ (p. 12). It should be noted that there is not just one structural arrangement of a composition that is the visually right one, but rather there are several locations within the spatial system of a visually right composition at which pictorial features could occur without disturbing the structural network and unitary effect of a composition created by a skilled artist (Locher et al., 1999).

Balance is believed by many to be the primary design principle guiding the distribution of pictorial elements within an artwork (Arnheim, 1988; Bouleau, 1980; Kandinsky, 1979; Locher et al., 1998, 2005). There is a considerable body of empirical evidence published over the years demonstrating that balance does, in fact, influence the immediate and sustained perception and evaluation of visual art (for review, see Locher, 1996). Balance is necessary because it unifies the structural elements of a visual display into a visually right, cohesive narrative statement, thereby creating the essential integrity or meaning of the work. Henri Matisse has put it this way: ‘In a picture, every part will be visible and will play a role conferred upon it by the artist, be it principal or secondary. All that is not useful in the picture is detrimental. A work of art must be harmonious in its entirety; for superfluous details would, in the mind of the beholder, encroach upon its essential elements’ (cited in Stangos, 1994, p. 24). A pictorial configuration is said to be balanced when its elements and their qualities are poised or organized about a balancing center so that they appear anchored or stable.

The simplest and most perceptually salient type of balance is the phenomenon symmetry. In this case one half of a composition appears as a mirror-image of the other about one of its principal axes (i.e. the vertical, horizontal, or diagonal axes); such an image is said to exhibit a formal type of balance called bilateral symmetry. Far more complex and interesting are objects and images whose elements are grouped or organized asymmetrically about a balancing center in such a way that their visual forces compensate one another. This type of equilibrium is often referred to as dynamic symmetry and is much more common in the visual arts than formal symmetry. The major stimulus features that contribute to a composition’s structural balance are: (i) the distribution of the perceived ‘weight’ of each pictorial component about the vertical, horizontal, and diagonal axes; (ii) cue directionality – principally with respect to left–right lateral organization; and (iii) the location of the areas of principal perceptual and semantic interest.

If a visually right composition has a very efficient structural organization, then it is reasonable to expect that its organization should be salient to both art-trained and art-naïve viewers. Locher and his colleagues have conducted a number of studies described below to test this assertion. In all of them, the art stimuli consisted of the same set of reproductions of original abstract and representational paintings by renowned artists whose structural organizations were deemed visually right by a group of artists, art historians, and designers. Additionally, two less well-organized versions of each original composition were created by experimentally manipulating with computer graphics programs the location of one or two pictorial elements deemed to be an integral part of the original’s balance structure. The artist’s signature was removed as a cue from each painting. Except for the alterations made, differences among the versions are indistinguishable either as computer screen images or as high-quality printed versions.¹

In one study by Locher (2003), undergraduate students who had no formal training in any of the visual arts examined three versions of each of four abstract and four representational works – an original balanced artwork and slightly and highly altered less-balanced versions of it. For each set of three versions, participants selected the one they considered to be a reproduction of the original painting and then described as many characteristics or properties of the picture as they could that contributed to their selection of it as the original, as well as any reactions to their choice. The artworks and position of the three versions relative to each other were presented in a different random order across the subjects. Locher found that participants, in general, were significantly more successful in discriminating between originals and the highly perturbed version than was expected by chance. As is typically the case for research in aesthetics, complex interaction effects were found among the variables studied. Specifically, hit rates for originals were highest for abstract works when a participant focused on their compositional style and form (as determined from recorded verbal reactions to a work) and highest for representational works when their content and realism were the focus of attention. These findings support the view that visually right compositions have efficient structural organizations that are salient to viewers who lack formal training in the arts.

If there is such a thing as a visually correct art composition then such compositions should be salient to viewers from different cultural backgrounds. Lega et al. (2003) conducted a cross-cultural study in which they recorded the preferences of educated and uneducated individuals living in Colombia, Brazil, Portugal, and the United States for one of the three versions of each artwork employed by Locher (2003). Although there was considerable variation in the rates of preference as a function of educational level and cultural background, preferences for the original balanced compositions were well above chance levels. These findings provide support once again for the notion of visual rightness, and the importance and salience of pictorial balance to the structural organization of visual art.

Another type of evidence that the visual rightness of a composition and its perceived beauty exist at a basic level and is ‘built into’ the perceptual system comes from a study by Watanabe (2009) that demonstrated that pigeons can be trained to discriminate the human concept of ‘good’ from ‘bad’ art. The stimuli in his study consisted of 15 ‘good’ and 15 ‘bad’ drawings by children (age range 9–11 years) selected from a larger sample of drawings classified for their beauty by 10 adults who unanimously agreed on the quality (good or bad) of the drawings selected for inclusion in the stimulus set. The drawings were representational in style and either of pastel or watercolor medium. The pigeons were trained using operant conditioning discriminative procedures to peck only at the ‘good’ pictures (at a 90% accuracy criterion). In a number of experiments, Watanabe observed that the pigeons responded to newly presented ‘good’ and ‘bad’ paintings differentially following training; they exhibited this ability when the original images were reduced in size and when one quarter of an image was occluded. However, discrimination performance reliably decreased when the stimuli were presented as grayscale images and when a mosaic effect was applied to the original images to disrupt their spatial frequency indicating that the pigeons used both color and pattern cues for their discrimination. Results were the same for both pastel and watercolor pictures. Watanabe’s findings demonstrate that pigeons have the ability to learn to discriminate stimulus properties of ‘good’ (visually right) paintings from ‘bad’ ones, according to a human concept of beauty. As mentioned, this finding supports the view that, although beauty may be a perceiver and a socially constructed concept, certain features or properties of visual art (e.g. symmetry, pictorial balance, and complexity) exist at a basic perceptual level. Whether humans and animals share similar aesthetic sensibility to these features remains to be determined.

3.2.3 Art on a Plate

A meal begins with the eye, with the visual appeal of the food on a plate, at least when a meal is a gastronomic occasion. This is most likely the reason so much time is spent by students in culinary institutes on the importance of presentation aesthetics (i.e. plating), and why there are thousands of sources including books, magazines, and Internet sites that describe and illustrate ways to use balance to create aesthetically pleasing presentations of food on a plate. There are, however, almost no empirical investigations that have assessed the validity of the claims made in these sources. One exception is a recent series of studies conducted by Zellner and her colleagues (Zellner et al., 2010, 2011) that investigated whether balance in the presentation of food on a plate affects the attractiveness of, the willingness to try, and the liking for the food presented. They hypothesized that balance in a culinary presentation would function in the same way it does when a viewer is experiencing a work of art.

The food stimuli in one study (Zellner et al., 2011) were a balanced and an unbalanced composition of the same foods. The balanced version consisted of a romaine lettuce leaf that was centered on the plate in a vertical orientation. A tablespoon of hummus was placed in the middle of the lettuce leaf and was surrounded by four pita chips, three grape tomatoes, three baby carrots, all evenly spaced symmetrically around the plate. The unbalanced version consisted of the same ingredients, but the chips, tomatoes, and carrots were placed asymmetrically about the hummus, and the entire collection of ingredients was positioned more in the bottom half of the plate. University student volunteers were presented with one of the plates in a laboratory setting and asked to rate the attractiveness of the food presentation, their willingness to taste the hummus, and, once tasted, how much they liked the hummus. Zellner et al. (2011) found that hummus in the balanced presentation was judged as tasting significantly better than the hummus in the unbalanced presentation although there was no significant difference in attractiveness ratings of the food or participants’ willingness to taste the food. Additional experiments revealed that the effect balance has on liking for the taste of food is mediated by the fact that balanced arrangements are perceived as neater (more visually right) than unbalanced arrangements, which in turn suggested to participants that more care was taken in their presentation (and possible preparation), whereas the items in the unbalanced presentation appear to be just ‘thrown’ on the plate. In another study in this series the researchers observed that the messiness of presentation significantly influenced the price that participants were willing to pay for the food; they were willing to pay more for neat presentations than messy ones.

3.2.4 Film

Film is the most popular visual art form worldwide and a very large empirical literature has emerged on the various antecedent factors that contribute to an aesthetic experience with film. These include studies of the contribution of viewer characteristics to the experience of film (e.g. Silvia and Berg, 2011) and the effects of variables such as a film’s budget, season of release, genre, attributes of the screenplay, ‘star power’ of the actors, etc., cinematic success as judged by film critics, movie awards, and box office performance (e.g. Simonton, 2009). Apropos to the topic of this section of the chapter, an artwork as stimulus, are investigations of the physical attributes of film, and how they engage viewers’ perceptual and cognitive processes. In this type of research, often called cinematics, data are derived directly from the art form, the stimulus, and not from observers’ reactions to it. A recent example of this type of study was conducted by Cutting et al. (2011) who developed a new physical measure for film called the visual acuity index (VAI). This measure captures the change in the amount of motion and movement in films.

Cutting et al. (2011) applied the VAI to a sample of 150 Hollywood-style films representing several genres that were released every five years from 1935 to 2005. They found a roughly linear increase in motion and movement across the time period measured, with adventure and action films showing the greatest change over time, and drama and comedy films exhibiting less of an upward tendency for increases in motion and movement. The duration of shot lengths was also observed to have generally decreased since 1935 and to be significantly inversely correlated with VAI across this period. According to Cutting et al. (2011), their findings ‘suggest that contemporary viewers of film have grown accustomed to, and desire to see, films with more visual activity than those that their parents and grandparents enjoyed’ (p. 120). The researchers propose a framework to address the issue of the extent to which this trend can continue given that viewers need relief (a refractory period) after being visually and cognitively challenged by heightened action especially as it approaches a level of visual chaos.

3.2.5 Neural Correlates of Aesthetic Judgments

Neuroscience has been shown for every area of the arts to be an important tool to help localize and connect the components of an aesthetic process with specific events in the brain. Some have argued that the future understanding of aesthetics, then, is to be found in neurology (e.g. Tallis, 2008). Massey (2009) cautions, however, that neurology is ‘of great value in exploring the “how” of aesthetic process, if not necessarily the “why” or the “what for”, or in helping decide whether one work of art is of greater value than another’ (p. 18). Neurology is, he argues, more effective in the analysis of component processes than in their synthesis. Neural data can reveal and localize the components of an aesthetic experience, but they are much less successful in enabling researchers, artists, and art theoreticians to understand how these components interact to produce the aesthetic experience or why one art work is perceived to be more beautiful than another.

An example of this type of research that relates to the topic of this section is a study conducted by Jacobsen et al. (2006) designed to isolate the neural correlates of aesthetic judgments of beauty using symmetrical and asymmetrical patterns previously shown to influence aesthetic judgments of visual images. Their stimuli consisted of black and white novel graphic patterns composed of different basic elements such as triangles, squares, and horizontal, vertical, or oblique bars of different sizes. These patterns were employed because they minimized as much as possible influences of attitudes or memory-related processes from the evaluative tasks performed. Half of the displays were symmetrical about either one or two symmetry axes and the other half were totally asymmetric; the patterns also varied in information content (i.e. complexity). Adult volunteers untrained in the visual arts performed either an evaluative aesthetic judgment task (beautiful or not) or a descriptive symmetry judgment (symmetric or not) following the onset of one of the stimulus patterns selected randomly. The neurological activity of each participant was recorded using functional magnetic resonance imaging (fMRI) as participants performed their task. A resting baseline control condition was also employed.

With respect to the imaging results, Jacobsen et al. (2006) found that both types of judgments relied on a set of brain areas that support high-level visuospatial analysis. Symmetry judgments had no significant influence on neural signal changes for symmetric versus asymmetric patterns. For aesthetic judgments, however, activation was additionally located in brain regions that have previously been reported for moral and social evaluative judgments about persons and actions and for the processing of introspective evaluation of one’s own thoughts and feelings. Additionally, beautiful judgments produced higher signal changes, and more viewing time than not-beautiful judgments. Of particular relevance to the present discussion is the finding that, when participants judged a pattern to be beautiful, not only were areas dominant in aesthetic judgments activated, but also the area activated during symmetry judgments showed enhanced activity. Jacobsen et al. (2006) point out that this neurological finding nicely parallels the behavior finding frequently reported in the literature that symmetry guides aesthetic judgments of beauty.

3.3.1 The Two Phases of an Aesthetic Experience with Pictorial Art

The aesthetic experience with pictorial art such as paintings occurs in two phases (Locher et al., 2007), which is generally believed to be the case when one encounters any visual stimulus. With the initial glance at a painting, the viewer spontaneously generates a global first impression (or gist) of it, which includes a sense of the work’s pictorial content, its overall structural organization, its semantic meaning, and an initial affective reaction to it. If the gist information perceived at first glance is deemed to have sufficient interest to the viewer, the second phase of the aesthetic experience ensues. This consists of focal exploration of the image in the form of eye fixations and movements directed to specific components of the composition to expand upon knowledge about the relationship between interesting pictorial features and their structural organization, to satisfy the viewer’s cognitive curiosity, and to develop his or her appreciation of the work. This focal exploration of the image continues until the observer decides that this information has been obtained. One can recognize these two phases of an aesthetic experience in visitors’ behaviors in a museum setting. As an individual moves from one artwork to another in a gallery, he or she makes a rapid decision based on one – or at most a few – glance(s) at a work that its contents are of very little or no interest and quickly moves on to the next work. When an artwork is initially found to be interesting and/or pleasing, the observer stops and spends time looking at it. As mentioned, whether one decides to ignore a painting or take time to ‘savor’ it, the aesthetic experience is driven by the complex interaction among artwork, observer, and context factors.²

3.3.2 Techniques Used to Study the Two Phases of an Aesthetic Experience with Pictorial Art

To examine the amount and type of information available to an observer from a visual display ‘at first glance’, art stimuli are flashed for 100 milliseconds (ms in what follows) or less (referred to as tachistoscopic viewing). The duration of a single fixation directed to a visual display is, on average, 300 ms. Thus, with just a 100-ms view of a picture, followed immediately by another stimulus display such as a field of lines to neurologically ‘mask’ the initial image content, the observer has access to just one view of the content of the pictorial field. What he or she ‘sees’ in the image, as indicated, for example, by the content of a viewer’s verbal response to the image, provides insights into the nature of the information and levels of perceptual organization available to a viewer at first glance.

A variety of techniques used to record the temporally unfolding, moment-by-moment focal exploration of and psychological responses to an artwork are described below. In the field of experimental aesthetics, as in many other areas of research interest, eye movement recording techniques have proven to be a very useful tool to reveal the perceptual and cognitive component processes that underlie an aesthetic experience with the visual arts. As an observer looks at a painting, for example, his or her eyes move in rapid jumps called saccades followed by pauses or fixations. In general, one’s eyes move three or four times per second during visual search; however, the number, duration, and location of the fixations used to scrutinize an art form are determined by the interaction among the many factors mentioned above. By superimposing a viewer’s eye movements over an image of a display he or she has examined one can map out the temporal location and sequence of fixations, called a scanpath, and identify the pictorial features of it that raised the viewer’s interest and invited exploration. A scanpath provides a graphic record of how information from an artwork was selected and processed by a viewer moment-by-moment across the time course of an aesthetic experience with it. Treisman (2006) cautions, however, that ‘the window of attention set by the parietal scan can take on different apertures, to encompass anything from a finely localized object to a global view of the surrounding scene’ (p. 4). Thus, scanpaths provide accurate information about fixations, but may not always lead to correct conclusions regarding the viewer’s focus of attention and conclusions concerning the underlying cognitive processes involved.

Locher et al. (2007) used these two techniques to investigate the nature of the processes that occur during each of the two phases of an aesthetic experience. The visual arts stimuli for both studies were eight paintings by renowned artists representing a range of artistic styles along a representational (Vermeer’s Young Woman with a Water Pitcher)–abstract (Klee’s Temple Gardens) continuum. In their first experiment university students with no training in the visual arts saw a brief (100 ms) flash presentation of each painting after which they immediately wrote five descriptions and/or impressions of the painting they would tell someone who had never seen the work in order to describe it to that person. The artworks were presented a second time for 100 ms each to the same participants whose task it was to rate the pleasingness of each work. In the second experiment a comparable group of volunteers was given unlimited time to view each painting and evaluate it for pleasingness. Their eye movements and verbal reactions to and thoughts about the painting were recorded simultaneously and continuously as they viewed each work. The think-aloud verbalization procedure employed is used in research to obtain insights into participants’ responses to a stimulus and the cognitive processes they use to arrive at their reactions.

To examine the content of the participants’ first impressions of the artworks following limited viewing in Experiment 1, the five statements about each painting made by each participant were classified in terms of six reactions that reflect a qualitative continuum of response ranging from attention to individual physical properties and pictorial elements of the composition (e.g. ‘The first thing I noticed was the cat’), to two or more elements described as a pictorial unit (e.g. ‘There is a woman reading a book’), to more holistic properties and characteristics of the compositions including their realism, beauty, expressiveness (‘The painting seems somewhat intimate’), and to style and form (‘The painting has many different contrasts in color and shape’). Locher et al. (2007) found that few of the participants’ initial reactions to the paintings or their additional four comments reflected attention to single compositional elements (2% and 4%, respectively). Rather the majority of all reactions to the artworks were evenly distributed among statements that reflect attention to a group of pictorial elements perceived as a compositional unit, to the expressiveness of the whole composition, or to its artistic style.

In the unlimited viewing time study, it was observed that participants began to describe holistic characteristics of the artworks of the type described above approximately two seconds after the onset of the picture. These reactions were based on the content of at most the first few glances at an artwork given that the average duration of a fixation in aesthetic research is approximately 300 ms. Furthermore, the distribution among the various types of responses did not change reliably following a viewer’s first full statement about a painting until he or she provided a pleasingness rating when finished looking at it (on average 30 seconds after onset). Additionally, the pleasingness ratings of the paintings obtained following the brief presentation correlated 0.73 with the ratings obtained from participants given unlimited viewing in the second experiment. This similarity in ratings suggests that the evaluation of an artwork’s pleasingness can be made rapidly, as is typically observed in a museum setting, as mentioned above. It must also be noted, however, that the average pleasingness ratings for the set of artworks for the limited and unlimited viewing time experiments were 4.59 and 6.17, respectively, on a 10-point scale, demonstrating that the paintings were evaluated as significantly more pleasing when observers had unlimited time to view them.

The eye-fixation pattern for each participant for each painting in Locher et al.’s (2007) second experiment was quantified by superimposing a 5 × 5 grid over the pictorial field of an image and measuring the percentage of total fixation time in each grid location during each of three time periods across the aesthetic episode. The periods consisted of the first 3 seconds of viewing, from 3 to 7 seconds of viewing, and the total viewing period. These temporal periods were utilized because it was observed that in almost all cases participants began to speak about the artwork between 2–3 seconds after it appeared on the screen and 98% of all initial reactions were completed within the first 7 seconds of visual exploration. It was found that observers’ initial reactions to an artwork, which were overwhelmingly holistic in nature, were generated from direct fixation of compositional features in approximately one-fourth (27% on average across the stimulus set) of the pictorial field. By the time observers completed their first statement about an artwork they had significantly expanded their coverage to 38% of the pictorial field. Coverage during the second focal exploration stage of processing increased to 46%, on average, which was not a significant increase over coverage at 8 seconds. Furthermore, the regions of the pictorial fields that received direct fixation were in the center of the composition and the outer regions of each composition were not directly fixated or did not receive sustained attention – a consistent finding in eye movement studies that used reproductions of artworks as stimuli. This seems to reflect the fact that pictorial information in the peripheral region of many compositions serves as a backdrop for the elements of principal interest that are typically located in the center of a composition (Arnheim, 1988). Furthermore, observers’ attention to the paintings during the second phase of the aesthetic episode was characterized by an even distribution of short- and long-dwell fixations. Previous research (e.g. Nodine et al., 1993) has shown that short-dwell fixations (durations of 300 ms or less) are used to scan the pictorial field to locate attractive, informative, interesting features. Periodically, when such features are discovered, long-dwell gazes (durations greater than 400 ms) are directed to such features to resolve questions of their meaning and significance to the overall compositional theme or to ‘savor’ particular elements.

Another example of how the study of eye movements provides insights into the interactive processes underlying aesthetic experience with visual art is the study conducted by Kapoula et al. (2009). They examined the nature of the interaction of stimulus-driven influences (the degree of abstractness of a painting) and cognitively driven influences (manipulation of information concerning a painting’s title) on the perception and aesthetic judgment of three cubist paintings by Fernand Léger. These included The Wedding (composed of fragments of human faces, limbs, and arbitrary pictorial fragments), The Alarm Clock (composed of pictorial fragments creating the perception of a person), and Contrast of Forms (consisting of forms and cylinders). Art-naïve adults explored each artwork in one of three conditions: either without knowing the work’s title (called the spontaneous condition by the researchers), with the instructions to invent a title for each painting (the active condition), or they were given the actual title of the work prior to viewing it (the driven condition). Participants’ eye movements were recorded as they viewed for an unlimited time each of the three paintings presented on a computer screen after which they were interviewed with regard to their perception and comprehension of each painting.

Kapoula et al. (2009) found both commonalities and differences in visual encoding of the artworks during the first and last 5 seconds of exploration across the three paintings and three conditions. Specifically, they observed that more fixations were directed to the center of each composition than to any other region of the pictorial field across the time course of exploration and fixations were more highly concentrated in the central area during the first 5 seconds than the last 5 seconds of exploration. The researchers suggest this finding demonstrates the important contribution that information in the central area of the compositions makes to the observers’ initial impression of the artworks. Additionally, it was found that participants tended to focus their gaze along the vertical axis of each painting about which the important structural components of each composition are located. Significant differences in exploratory behavior were also observed as a function of the title manipulation and specific artwork. Fixation durations were different, averaged across all paintings, among the three title conditions; durations were longest in the driven condition, shorter in the active condition, and shortest in the spontaneous condition. Kapoula et al. (2009) suggest that this is due to the different levels of cognitive activity required by the three tasks. For example, they speculate that being given the actual title of a painting in the driven condition caused viewers to implement a search strategy to fit the contents of the painting to its title that engaged deeper levels of semantic analysis, and correspondingly a greater number of long-dwell fixations, than utilized in the other two task conditions.

The specific pictorial content of each painting also influenced in a differential way participants’ exploration of the stimuli. For example, The Wedding ‘produced’ smaller saccade sizes (i.e. the distances between fixation locations) in the active and driven conditions than was the case for the other two paintings, presumably due to visual aspects of the compositions (i.e. the high density of small pictorial fragments) and to the detailed semantic analysis required by the many real human faces and limbs contained in this painting. Collectively, Kapoula et al. (2009) findings demonstrate the complex interaction of bottom-up (stimulus-driven) and top-down (cognitively driven) influences on visual exploration of paintings during an aesthetic episode.

3.3.3 Pupillometrics

Another non-verbal viewer reaction to art is measured and studied by examining changes in the size of his or her pupils across an aesthetic experience. This psychophysical approach, known as pupillometrics, has been used extensively for many years to measure neurologically the effectiveness of different types of social attitudes, marketing strategies, and clinical and therapeutic procedures, and it has been shown to be reliably sensitive to cognitive aspects of information processing as well as affective components of emotional processing (for a review of this literature, see Hess, 1975). Thus, as a study by Kuchinke et al. (2009) illustrates, pupillometrics has the potential to provide insights into the affective processes involved in art appreciation. They investigated the relationship between the ease of processing (i.e. processing fluency) of the content of abstract cubistic art and positive aesthetic emotion. The task for the art-naïve university participants was to press a mouse button at the moment they recognized a ‘realistic’ object depicted in an abstract painting. The paintings differed across three levels of increasing content accessibility (low, medium, and high). For example, the face of a woman in Picasso’s Femme et Pot de Moutarde is readily visible in this high processing fluency painting, whereas the boats in George Braque’s Fishing Boats are low in visual accessibility.

The time needed to recognize a depicted object and the viewer’s pupil size following the point of recognition were recorded. Subjects rated each painting’s familiarity, complexity, and their preference for it following the initial stage of the experiment. Kuchinke et al. (2009) found that the time to recognize a depicted object was shorter for high processing fluency paintings, which were also rated highest for preference. Higher processing fluency was also associated with large pupil dilation following the point of recognition of an object. Furthermore, pupil dilation following a subject’s behavioral response was positively related to individually rated preferences of the abstract paintings. The researchers interpret their findings as supporting the ‘hedonic fluency model’, which predicts higher processing fluency of paintings being associated with positive aesthetic affect.

3.3.4 Personal Data Assistants

Recent developments in hand-held computers commonly known as personal data assistants (PDAs) coupled with wireless technology have made it possible to record continuous psychological responses from individual audience members viewing temporally unfolding live dance, theater, and music performance in naturalistic settings. The PDAs now available are small, portable, relatively inconspicuous, programmable, and make it possible for the data collected to be collated and time-stamped for synchronization purposes with action on the stage. The participants in such studies move a hand held pen-like stylus or a joystick around a PDA screen to generate input to the PDA. One of the first studies to use this technology was conducted by Stevens et al. (2009) who developed a portable audience response facility (pARF), a modified PDA system, to collect responses from audience members during two live performances of dance works. The researchers’ goal was to investigate audience perceptions of structures and expressions in the dances and to compare these to the choreographer’s descriptions of each work. Continuous responses were measured in the first study from participants who had no experience in contemporary dance as they viewed a live performance of a 60-minute contemporary dance piece entitled Fine Line Terrain – an emotionally charged dance in which five dancers navigate through a set consisting of a dynamic architecture of many white lines dissecting black space. Participants in the second study were accustomed to attending live dance performances and reported enjoying watching dance. They viewed a full-length live work; however, the researchers recorded PDA responses from a single section of the first act titled Earth: Silent Heartbeat, which varies in the amount of activity on stage reflecting the themes ‘struggle’, ‘achievement’, ‘reflectiveness’, and ‘loss’.

Participants in both studies moved the stylus around the PDA screen while watching the performance to rate the emotion that they judged was being expressed by the entire performance at any given time (i.e. by the dancers, the choreography, and the soundscape). They were given a practice session prior to the performance to acquaint them with the use of the system. The PDA recorded simultaneously two dimensions of expressed emotion as the performance took place: a valence scale (positive–negative) and an arousal scale (aroused–sleepy). The continuous response data were mapped onto choreographic notes from the choreographers of the two ballets that indicated the key moments and intended structural and musical changes in the pieces. Stevens et al. (2009) present a detailed comparison of the connection between these variables across the time course of a segment of the ballet. In general, they found numerous examples where the choreographers’ intention and the audiences’ emotional responses were congruent (approximately 80% of the time). For example, greater activity by the dancers, such as acrobatics, and changes in musical material were associated with changes in arousal and emotional responses, demonstrating a connection between audience emotional responses and the features of the dance and music as well as the expressive aspects of the performance. It should be noted that participants reported the PDA to be very user-friendly and that it did not distract them from the performance – a potential drawback for use of such a system during live performances. Thus, Stevens et al. (2009) findings enable choreographers, dancers, and researchers to see for the first time the actual responses to dance works made in real-time during a performance.

3.3.5 Computerized Dynamic Posturography

Another approach to measure physiologically a viewer’s reaction to art temporally is computerized dynamic posturography, which is a non-invasive system that quantifies body sway. The body is not perfectly still when one is looking at an object with the body in an upright quiet stance. Rather, continuous postural adjustments are made that involve the dynamic coupling of information from the visual, vestibular, proprioceptive and somatosensory, and central systems to control and maintain posture and balance in such circumstances. Changes in center of pressure (CoP) of the body serve as the measure of postural responses to visual stimuli. CoP data are typically recorded as a viewer stands on a fixed instrumented platform, or force plate, which is connected to sensitive detectors capable of measuring changes in the force exerted on the surfaces of the two feet. Changes in excursion of the CoP, instantaneously adjusted for a person’s height and weight and distance from the source of stimulation, are recorded in both the anterior–posterior and medial–lateral directions. The first commercially available system to measure body movement became available in 1986 for use in the diagnosis of balance disorders and in physical therapy and postural re-education.

Posturography has been used to study viewers’ reactions to emotionally charged pictorial images. For example, Hillman et al. (2004) collected postural adjustment data from male and female undergraduate students as they observed pictures depicting three types of affect: pleasant pictures that included scenes of erotica and families, neutral pictures depicting neutral faces and household objects, and unpleasant pictures depicting scenes of attacking animals and disfigured bodies. A subjective report of the affective valence and arousal of each picture, and a startle eye-blink reflex to each image were collected from the same participants during a second session. The researchers observed sex differences in postural responses to unpleasant pictures, but no effect for pleasant and neutral picture contents. Specifically, males exhibited increased movement in the anterior direction (approach) for unpleasant pictures, whereas females exhibited increased postural movements in the posterior (withdrawal) direction. These findings provide support for the existence of a relationship between affective reactivity and motivated approach–withdrawal behavior and suggest that females react with greater defensive motivation than males. Furthermore, participants rated pleasant pictures higher in valence and they exhibited smaller startle responses to these images as compared to unpleasant pictures. Clearly the study of postural sway has great potential to measure temporally viewers’ affective reactions to the contents of emotion-eliciting art forms, but this author is not aware of any such research findings being reported in the literature to date.

It has been shown, however, that the study of postural adjustments of the body to the pictorial elements and overall compositional organization of artworks can provide a physiological measure of a viewer’s reactions to these stimulus properties. For example, Nather et al. (2010) asked university students untrained in the visual arts and classical dance to view photographic images of two sculptures of ballerinas created by Degas. One ballerina was in a static position and the other was taking a large step that appeared to place the ballerina in a dynamic state of potential imbalance. The task of participants, who were standing on a force platform, was to observe each image and estimate its presentation time duration. Nather et al. found that participants moved their bodies much more when they observed the dancing ballerina, demonstrating that images of body movement internally generate unconscious body oscillations.

In another posturography study, Kapoula et al. (2011) examined the effects of pictorial depth on body sway and found a significantly greater increase in sway when observers fixated the principal background area versus the foreground of Piero della Francesca’s painting Annunciation. Additionally, Kapoula et al. observed significant differences in postural sway when viewers examined two abstract paintings by the artist Maria Elena Vieira Da Silva (Egypt and O Quarto Cinzento); greater postural sway was recorded for the painting O Quarto Cinzento, which contains greater intensity of visual depth cues. The effect was lessened when participants viewed cubist transformations of the same paintings in which pictorial depth cues were experimentally neutralized.

The visual impression of motion in an artwork is created indirectly from the static, two-dimensional ingredients that make up the work’s pictorial elements and organization. Recently, this author and his colleague (Locher and Kapoula, 2012) reported evidence using posturography that the depicted motion cues and the structural organization in paintings have the ability to physically move a viewer during an aesthetic experience with art, and that the neural mechanisms responsible for body sway are differentially sensitive to the strength and lateral organization of implied motion cues in paintings. Our stimuli consisted of a pair of ‘matching’ compositions by Monet that depict movement produced by the wind and the reversed-view or mirror-image versions of the two originals. The paintings, entitled in English Study of a Figure Outdoors: Facing Right and Study of a Figure Outdoors: Facing Left (the original titles are Essai de Figure en plein air: Femme à l’ombrelle tournée ver la droite and Essai de Figure en plein air: Femme à l’ombrelle tournée ver la gauche, respectively), are very similar in overall structural content and organization, color, and size. In both works the wind is depicted as blowing from right to left, as indicated by the depicted movement of clouds, the woman’s dress and scarf, vegetation, etc. However, in one version the woman, who is standing at the top of a grassy knoll, is facing into the wind and in the other version she has her back to the wind. Image cues suggest a stronger perceived effect of the wind in the painting in which the woman is facing to the artist’s left.

In addition to investigating the influence of pictorial cues in the originals on body sway, the similarity of the compositional properties of the two paintings made it possible to determine if differences in salience of implied motion cues would occur between reversed-view or mirror-image versions of the originals. Studies, the details of which need not concern us here, have demonstrated that viewers tend to prefer paintings depicting left-to-right organization believed to be due to an attentional bias to the left visual field based on lateralization of hemispheric functions (e.g. Mead and McLaughlin, 1992).

In our study, university students with no training in the visual arts stood on a force pad that provided measures of postural body sway as they viewed actual-size images of the paintings in a counterbalance order following a baseline condition. Analyses revealed differential effects for body sway measures between the two paintings as a function of their depicted depth cues, overall structural arrangement, and lateral organization. Specifically, the surface area of body sway, the variance of the speed of body sway, and the standard deviation of the body sway along the medial axis were each found to be greater, on average, when the painting Study of a Figure Outdoors: Facing Left was viewed in mirror orientation than when it was projected normally. This indicates that the mirror image of this picture was more destabilizing than when it was seen in the normal orientation. Additionally, participants who viewed this painting in either its normal or mirror orientation exhibited a larger surface of body sway than when they experienced the baseline condition whereas participants who viewed Study of a Figure Outdoors: Facing Right in both normal and mirror orientations exhibited less surface area sway than occurred for their baseline condition. These findings suggest that, compared to their baseline conditions, the compositional arrangement of painting 1 (in which the depicted force of the wind was stronger) seen in both orientations was more destabilizing than the organizational structure of painting 2 seen in both orientations. Results provide evidence that people are physically moved when viewing art and that the mechanisms responsible for postural stability are differentially sensitive to depicted motion cues and their lateral organization in paintings.

3.4.1 The Influence of Formal Training in Art on Aesthetic Phenomena

The effects of training in an art form on aesthetic phenomena, such as those observed by Axelsson (2007), have been explained by a number of recent theories, including Axelsson’s Scheme Theory, Silvia’s (2005) Appraisal Theory, and the Processing Fluency Model of Reber et al. (2004). Underlying these and other explanations of the importance of expertise in art, and for that matter expertise in most other domains, are the often observed findings (Bruning et al., 2004) that experts, compared to non-experts, have greater domain and procedural knowledge related to their field of expertise that is highly inter-related in their knowledge base. This enables them to represent problems at a ‘deeper’ level of processing, to better monitor their performance, and to do so more quickly than non-experts. These differences in perceptual/cognitive processes as a function of expertise are illustrated nicely by the findings of a study conducted by Locher et al. (2008), described below, that examined the contribution of expertise and induced positive affect on an aesthetic evaluative task.

There is a large body of research that demonstrates that persons in whom positive affect has been induced externally in any of a number of simple ways, such as receiving a small unexpected gift of candy before performing a task, are likely to adopt more creative, constructive, and open thinking styles, to use broader cognitive categories, and show greater cognitive and behavioral flexibility across a wide range of tasks and settings. Research by Locher et al. (2008) has demonstrated that aesthetic responses to new product designs can also be influenced by induced positive affect. Specifically, they had male university students trained and untrained in principles of design theory rate six newly designed digital cameras for their aesthetic appeal. Half of the participants received a small bag of candy prior to completing the task as a token of the researchers’ appreciation for their participation in the study; the control groups did not receive candy. Each camera was presented in a different random order on a turntable that participants used to rotate the camera to obtain all desired views of it; participants were not permitted to handle the cameras. Additionally a think-aloud procedure was employed that required participants to verbalize their thoughts as they completed the task. The verbal records provided insights into the processing strategies responsible for the anticipated positive affect infusion in the form of the number and affective quality (positive, negative, neutral) of participants’ reactions to the cameras’ features. A record of the entire session was made with a digital camera.

Locher et al. (2008) found that, as anticipated, participants given candy rated the set of cameras as significantly more appealing, on average, than did individuals who were not given candy. Verbal protocol analyses revealed that the cognitive processing styles responsible for the observed influence of positive affect on the viewers’ judgments were differentially influenced by training. Specifically, the remarks of unsophisticated participants who received candy contained significantly fewer reactions overall than did those who did not receive candy; however, their reactions were significantly more positive in nature. This finding is consistent with previous studies that show that people in a positive mood tend to cease goal-directed behavior because they interpret a positive mood as reflecting their level of goal attainment for the task. On the other hand, design students who received candy generated more reactions to the cameras than design students who did not and these were predominately neutral in type for both groups. This finding led the researchers to suggest that the focus of the trained individuals was on the structural features of the cameras, and that they used viewing/processing time to select stimulus information and relate it to their domain-specific knowledge about design to generate an evaluative response. Locher et al. speculate that being in a positive mood may have facilitated seeing more aspects of a camera’s design features and, hence, result in a more positive evaluation of its appeal. Note that without the data made available by the think-aloud procedure, the complex interaction of stimulus, subject, and task factors on an aesthetic interaction would not have been possible. This study demonstrates the value of the little used think-aloud procedure in experimental aesthetics research.

3.4.2 Aesthetic Fluency

Most researchers who have evaluated the contribution of expertise to aesthetics have contrasted the reactions to art forms between groups of novices and experts, often between those who self-report no formal education or training in the art form studied or art-related fields with those who have had training. There are two important limitations to this methodology. (i) Sophistication in the arts has been shown to develop along a continuum (see stage models of art appreciation: Parsons, 1987; Housen, 1992) and should not, therefore, be treated as a nominal variable with two levels as is typically the case. (ii) Art expertise is not simply acquired by direct instruction, but is also acquired through experience with art in museums, in galleries, in school, by reading books, and increasingly on the Internet.

Smith and Smith (2006) introduced the notion of aesthetic fluency to address these limitations. They define aesthetic fluency as the knowledge base that an individual has about art and aspects of life closely related to art that facilitates aesthetic experience in individuals as they develop – a notion consistent with stage models of art appreciation. The Smiths developed the Aesthetic Fluency Scale to measure this characteristic of a perceiver. The scale, which provides a continuous measure of sophistication, was created from the responses of a large sample of visitors to the New York Metropolitan Museum of Art to a survey that asked them to rate how much they knew about five ideas in art (Fauvism, Egyptian Funerary Stelae, Impressionism, Chinese scrolls, and Abstract Expressionism) and about five artists (Mary Cassatt, Isamu Noguchi, John Singer Sargent, Alessandro Botticelli, and Gian Lorenzo Bernini). In addition, participants responded to a number of demographic questions concerning their museum visitation history, overall educational level, and training in art history. Analyses of the full set of data revealed that aesthetic fluency, as measured by the Aesthetic Fluency Scale, develops gradually and broadly across areas of art as evidenced by the observed continuous distribution of knowledge on the scale’s 10 items across the sample of museum goers in the study. Additionally, it was found that aesthetic fluency is greatly influenced in a positive way by frequency of museum visitations, age, and training in art history (but not general educational level). Furthermore, Silvia (2007) found that university students’ levels of aesthetic fluency were highly correlated with the Openness to Experience scale of the Big Five personality dimensions measured by the NEO Five Factor Inventory, but not to the Extraversion, Neuroticism, Agreeableness, or Conscientiousness scales. People high in aesthetic fluency were not generally found by Silvia to be smarter when assessed by measures of fluid intelligence, which represents general ability and achievement. Thus, use of the Aesthetic Fluency Scale in future studies of the contribution of art sophistication to the aesthetic experience with visual arts addresses limitations of previous studies in the field by providing researchers with a continuous quantitative measure (derived from the scale’s Likert rating format) of an individual’s knowledge about the arts rather than his or her informal self-reported level of sophistication on a nominal variable with two levels (e.g. trained–untrained, sophisticated–naïve).

3.4.3 Viewer Personality Traits Influence Aesthetic Reactions and Preferences

In addition to sophistication in the arts, several basic components of personality have consistently been related to aesthetic reactions and preferences. Feist and Brady (2004) report that the most consistent and robust findings of studies focused on personality characteristics of observers that determine their aesthetic responses to art show that one’s cognitive, behavioral, and dispositional flexibility, experience seeking, and openness to experience, are correlated with various aspects of aesthetic preference. They added to this body of knowledge with a study in which they sought to identify the psychological and social attitude profile of college students who prefer abstract art – a generally unappreciated form of art. Participants completed two openness-to-experience scales and an experience-seeking scale; they also responded to questions relating to tolerance of substance abuse and political liberalism. Additionally, participants provided demographic information pertaining to their age, gender, race, income, their class and major in school, and the religious and political affiliation of their parents. They were also given 5 seconds to rate their preference on a nine-point scale (anchored at ‘do not like at all’ to ‘like very much’) for each of 15 realistic, 15 abstract, and 15 ambiguous art forms presented in a random order.

Feist and Brady (2004) report that, overall, participants preferred realistic to abstract art and this effect was most pronounced for individuals low in openness to experience. Open participants, on the other hand, preferred every style of art viewed and their preference increased as the art became more complex. Additionally, those individuals who held attitudes that were more tolerant of drug use and political liberalism preferred abstract art the most. None of the demographic variables were significantly associated with preferences for abstract art. Feist and Brady conclude that, consistent with previous findings, openness to experience is the personality trait that is most consistently associated with aesthetic appreciation, even if art is very abstract. This finding was replicated recently for museum goers by Mastandrea et al. (2009) who observed that visitors to the Braschi Museum for Ancient Art and the National Gallery of Modern and Contemporary Art in Rome achieved high scores on the Openness to Experience personality measure of the NEO Five Factor Inventory, with no significant difference between the two groups of visitors on this measure. Not surprisingly, visitors to the modern and contemporary art collection scored significantly higher on a sensation-seeking measure than did those who viewed the collections of ancient art.

3.4.4 Sample Size

As is the case for all areas of the behavioral sciences, researchers working in the field of experimental aesthetics typically use data collected from a sample or subset of individuals to draw inferences about the behavior of the population of individuals from which the sample was drawn. A population is the set of all members of a specified group, such as all museum visitors, or more specifically, all visitors to the New York Metropolitan Museum of Art. It is imperative, therefore, that the sample employed in a study be truly representative of its population if the findings are to be valid (i.e. generalizable to the population being studied). Defining a population is a particularly difficult task in the field of experimental aesthetics because of the many viewer characteristics discussed in this and previous sections that influence in an interactive fashion individuals’ personal taste for art.

A problem that must be faced by researchers is to determine the size of the sample necessary to attain the goal of a study that can be considered valid (a detailed discussion of issues related to sample size and sampling procedures can be found in Borg and Gall, 1989, and in most texts dealing with research design). The rough-and-ready rule is to use the largest number of subjects possible. In most research projects, however, financial and practical restrictions limit the number of subjects that can be included in a single study. Generally, it is considered desirable to have a minimum of 15 cases in each group compared in an experimental study, 30 cases in correlational research, and at least 100 subjects in survey research. (Statistical procedures exist for estimating the sample size needed to achieve a desired level of accuracy of outcome for a given study.)

Researchers who conduct studies of the nature of an aesthetic experience with some form of art tend to employ samples that are at least as large as these recommendations, if not somewhat larger. This is the case for the studies described in this section and throughout the chapter. For example, participants in Locher et al.’s (2008) experiment included 20 university students trained in principles of design theory and 20 students who had no such training. Smith and Smith (2006) surveyed 400 individuals who were visiting the New York Metropolitan Museum of Art to develop their aesthetic fluency measure and Mastandrea et al. (2009) obtained questionnaire data from 191 visitors to either an ancient or modern art museum to examine the relationship between museum goers’ visitor experience and their personality characteristics. Feist and Brady (2004) collected data from 104 university students to identify the personality characteristics of observers that determine their aesthetic responses to art. Again, these are sample sizes that are typical of research conducted in all areas of the behavioral sciences.

3.5.1 Audio Guide Augmentation of the Museum Experience

Research shows that most art museum visitors like to have some organizing schematic information presented with works of art by these methods, but not to be overwhelmed with information, and they also want freedom to interpret the art as they will (see reviews in Tinio et al., 2010; Smith and Tinio, 2010). Furthermore, visitors also desire a balance between the physical organization and ‘mental’ structure of a gallery or exhibition and their freedom to explore and examine artworks and concepts according to their own organizing structure or notions about the art. The impact of these factors changes according to the nature of a work of art (e.g. abstract versus representational), the amount and type of information presented, and the characteristics and aesthetic fluency of a viewer. In the past, this complex interaction of factors could not be taken into consideration by a museum, and a museum experience was, in a sense, structured for the ‘average visitor’. Recent technological advances with hand-held computers have made it possible for museums to organize audio guide augmented tours in such a way that they provide each visitor with personalized choices in the amount, types, and depth of information they receive about a work, as well as the freedom to navigate a collection of artworks as they choose. Thus, audio tours have the potential to strike a balance between freedom and structure in the visit of individual museum goers.

To date, there is relatively little empirical research on the contribution of audio augmentation to the museum experience using the new technologies available for such tours. A notable exception is one of a series of studies conducted by Smith et al. (2004) and Smith and Tinio (2010) at the Whitney Museum of American Art in New York. This study examined the impact of the museum’s approach to providing audio tour information in a controlled in situ experimental fashion. Museum visitor volunteers completed surveys after experiencing one of four works of art (two sculptures two paintings) under one of three treatment conditions. The two paintings were Arshile Gorky’s The Artist and His Mother and George Bellows’ Dempsey and Firpo, and the two sculptures were Yayoi Kusama’s Accumulation and Chris Burden’s America’s Darker Moments. The Kusama sculpture is a white chair that has been covered with dozens of phallic-shaped, white stuffed bags. Burden’s work consists of five separate pieces that represent events in American history (e.g. events at My Lai, Kent State, etc.) using small metal sculptures. The three treatment conditions were: standard (the museum’s standard label that consisted of only the work’s title, artist, date, and acquisition information), label (a label written expressly for the study to present typical art historical and stylistic information), and audio (information specific to each work and its artist was presented³). Data for several dependent measures evaluating participants’ experience with each artwork were obtained; however, space permits the presentation of only a few general findings here.

With respect to the amount of time people spent in front of an artwork, statistically significant effects were observed for both object and condition. Specifically, the audio condition caused people to look at the works longer than was the case for the other two conditions. It was also found that participants said they would like to see more of the works by a given artist when his work was experienced in the audio condition than when it was seen in the standard (no information) condition. This was especially the case for the Kusama and Burden sculptures, which are the most difficult to understand and probably require more interpretive help to be appreciated. The specific information presented in the audio guide for these works provided a cognitive schema or structure for looking at and understanding these works.

One most interesting negative influence of the audio guide approach was that it appears to focus attention on only certain features and aspects of a work, and away from others. Participants were asked a question about something that was not included in the audio stop. For example, those who saw the Bellows work were asked about the color of the ceiling in the painting. It was found that with each of the four artworks, participants in the standard and label conditions were better able to answer questions about the works than people in the audio condition. Smith et al. (2004) suggest that attention was directed toward certain aspects of the work, and away from others, by the information provided in an audio stop. Thus, too much structure is not optimal. Clearly additional research is necessary to identify the observer and artwork factors that must be incorporated in an observer-specific interactive fashion to make audio guide augmentation a more successful technology as it continues to advance rapidly from a technological perspective.

3.5.2 The Influence of Structural and Contextual Factors on an Aesthetic Experience in a Museum Setting

In recent years survey research conducted in museums has revealed that social contexts within a museum also influence the way visitors engage with the art and how much they learn and remember about the contents of the collection or special exhibitions. To organize a museum’s holdings in a way to foster visitor learning experiences, Chang (2006, p. 181) notes that a museum must know who is visiting, why they are visiting and with whom, what they do and see in the museum, and how all of these factors interact and inter-relate. For example, visitor behaviors are linked to their personal agendas for the visit (i.e. discovery, enjoyment, relaxation, socializing). Ballantyne and Packer (2005) reported that the majority (frequently reported as high as 85%) of adult visitors to art museums attend in the company of a partner, friend, or family group. Interestingly, nearly 50% of couples attending a museum do not talk to each other while viewing the art and experience it in very much the same way as those who visit alone. Other studies show that those who visit alone have a higher need for cognition than those who visit in groups, are much more likely to use audio guides, spend more time reading labels, and demonstrate greater learning.

The art styles of the collection housed within a museum also attract different types of visitors in terms of the motivation for their visit. Mastandrea et al. (2007) report that visitors to the ancient art collection housed in the Museum Borghese of Rome were intent upon learning about and understanding its art and history, whereas visitors to the contemporary art collection housed in the Peggy Guggenheim Collection of Venice took an emotional and pleasure-seeking approach to their experience.

Art museums themselves are imbued by the public with high social status because they contain artifacts of high social value. It is believed that societal notions about the value of ‘high art’ lend a certain positive aura to a museum experience, although this author knows of no study that has attempted to determine just how much additional positive valence the museum itself adds to the aesthetics of the artifacts within. Supporting this view is the finding by Mastandrea et al. (2009) that the number one reason given by visitors for their attendance at the National Gallery of Modern and Contemporary Art in Rome was to see artworks in the original. There is an interesting paradox associated with the social context of a museum demonstrating the interactive nature of contributing factors to an aesthetic experience. Chang (2006) has observed that museums’ social status reinforces feelings of exclusion for some individuals while strengthening feelings of belonging by others. What keeps non-visitors from art museums, even on days when there is free entry, are their perceptions that museums are formal and formidable places that are for the upper-class who are knowledgeable about art, which, based on demographic studies of museum visitors, tends to be the case. Non-visitors see art museums as inaccessible to themselves because they feel they do not have sufficient background of the type they think necessary to interact with high art.

A research project that has made an important contribution to the understanding of the complex interaction of the factors that contribute to a museum visitor’s experience with artworks is the eMotion: Mapping Museum Experience study (Tschacher et al., 2012 and www.mapping-museum-experience.com). The study, which used innovative technical apparatuses, was designed to examine how the museum context, the art objects, curatorship, and the observer affect his or her behavior and experiences. The large trans-disciplinary team who conducted the research project at the Kunstmuseum St Gallen in Switzerland consisted of collaborators with expertise in art psychology, art sociology, art theory, museum visitor studies, curatorial practice, industrial design, and technical programming. Visitors who volunteered to take part in the study were invited to wear an electronic glove containing signaling sensors through which their locomotion, heart rate, and skin conductance were continuously recorded. These made possible the precise quantitative measurement of the path each individual took through the museum, his or her speed, the time spent in front of each artwork, and when and how strong an emotional arousal and a cognitive arousal occurred. After the visit, a participant’s aesthetic and emotional responses to selected artworks were assessed using questionnaires displayed on a computer terminal. A detailed description of the findings obtained by Tschacher et al. from their examination of the complete dataset is beyond the scope of this chapter. In general, they found that an observer’s aesthetic-emotional responses are statistically grounded in his or her physiology during the perception of an artwork in an ecologically valid environment – the art gallery. Another new aspect of this study was that participants had the opportunity to view moments of their own course through the eMotion installation played back through cartographic representation. This procedure provided self-insight into their behaviors and enabled them to reflect on the nature of their museum experience.

3.5.3 Is it the Case That When Viewing Visual Art, There is ‘Nothing Like the Original’?

Original works of art are almost always experienced within a museum or some sort of gallery or exhibition space. It is important for the sake of ecological validity, therefore, that investigations into the nature of an aesthetic experience with art be conducted in a museum setting. For practical reasons, researchers who investigate empirically this issue rarely have the opportunity to use original artworks as their art stimuli. Rather, their subjects see the art in a laboratory setting in some form of reproduction, either as printed pictures (books, posters, and postcards), slide-projected images, or, increasingly, as images on a computer screen. This procedure, which has been the case for almost all of the experimental studies described thus far, raises the question of the comparability of the findings of experimental aesthetics research to the responses of viewers in the presence of original art in a museum.

Given the widespread use of reproductions of art in aesthetics research, one would suspect that many studies have examined what is lost, or possibly gained, when an observer interacts with an original by a renowned artist in a museum as compared with reproductions of it in different image formats. In fact, there are almost no such studies reported in the literature since the new imaging technologies emerged that produce high-quality digital images of original art that seem to capture somewhat faithfully the physical qualities of art. One exception is the study conducted by Locher et al. (2001) at the New York Metropolitan Museum of Art. They examined the influence of image format on the perception and evaluation of pictorial and aesthetic qualities of paintings viewed in one of three different formats within the museum: either the originals were seen in the galleries, as slide-projected images, or as images on a computer screen. Subjects in the study were art sophisticated or naïve adult museum visitors who volunteered to participate. They rated each of nine paintings by renowned artists (e.g. Bruegel, El Greco, Rembrandt, and Vermeer) under one of the three format conditions on 16 measures of physical and structural characteristics, aesthetic qualities, and novelty of content. In a follow-up study, Locher and Dolese (2004) had art-trained and naïve university students perform the same task with postcard images of the nine paintings in a laboratory setting.

Results of the two studies revealed that ratings of the adjective pairs that assessed qualitative stimulus properties of the compositions (items: symmetrical–asymmetrical, homogeneous–heterogeneous, patterned–random, continuous–intermittent) and quantitative features (items: simple–complex, crowded–uncrowded, homogeneous–heterogeneous) were very similar across the original and three reproduction formats for both sophisticated and naïve observers. Moreover, sophisticated visitors consistently rated paintings across all formats as more complex, asymmetrical, varied, and contrasting than did naïve visitors. Thus, with respect to the physical and structural elements of the art, the four presentation formats exhibited what Locher et al. (2001) call ‘pictorial sameness’. On the other hand, in terms of the visitors’ evaluative judgments, the majority of the artworks studied (those by Chardin, Christus, Giotto, Rembrandt, van Eyck, and Vermeer) were rated significantly more pleasant, interesting, and surprising in the original format than in reproduction by all observers; ratings for the artworks by the sophisticated observers were consistently higher than those of the naïve observers across formats. These findings suggest that when it comes to experiencing the pleasure of art, the often heard adage ‘there is nothing like the original’ may in fact be true. As one would expect, differences in reactions to the artworks among the paintings were seen. Differences in the ratings of the pleasingness and interest across formats for the works by Bruegel, El Greco, and van Ruisdael demonstrate that much additional research into the influence of format is need to identify the characteristics of paintings that contribute to the hedonic value of a composition in the original compared to the reproductions of it that are typically seen by the public.