The Philosophy of Art and Aesthetics, Psychology, and Neuroscience

Alexander Baumgarten, 1735: Baumgarten used the phrase episteme aisthetike to mean the science of what is sensed and imagined, in contrast to the science of what is known through rational thought. In his incomplete Aesthetica, Baumgarten further defined the terms as follows: “Aesthetics (the theory of the liberal arts, the logic of the lower capacities of cognition [gnoseologia inferior], the art of thinking beautifully, the art of the analogon rationis) is the science of sensible cognition.” Since then, philosophers started exploring topics of:

• judgments of beauty: faculties of the judging subject and the qualities judged in the object
• taste and preference
• role of imagination
• aesthetic and morality
• experience of artwork

only with the advent of cognitive science in the late 20th century that Baumgarten’s vision of aesthetics as a science of sensory cognition, including the study of our engagement with artworks, has become truly possible.

The Viability of Aesthetic as a Science

• Logical Positivists: preferred to exclude anything that involves subjective feeling from the general domain of knowledge. Science could not, in principle, have any bearing on questions about the nature of art, beauty, or aesthetic appreciation.
• analytic philosophers have been generally skeptical about the relevance of psychology to the philosophy of art and aesthetics.
  • Ludwig Wittgenstein’s Lectures in Aesthetics：
  • George Dickie’s “Is Psychology Relevant to Aesthetics”
    • when we answer aesthetic questions with data concerning the behavioral responses of participants to particular artworks we are committing a category mistake—we confuse a logical problem for a scientific one: what matters in the case of our responses to music is whether the formal composition of the work—in this case, the syntactic features of compositions that constitute a musical language is adequate for the attribution of the aesthetic feature in question. And this, Dickie argues, is a matter of convention rather than psychological causation—a matter of the rules of the particular language game appropriate to the production and appreciation of the category of music used as the stimulus. Questions about the nature of art and the character of aesthetic experience are prior logical questions about the use of a language.
  • Refute to Dickie’s arguments: recent imaging studies involving adjective matching and preference orderings support Dickie’s observations but challenge his assessment of the relevance of psychology to aesthetics: correlations among Calvo-Merino’ imaging results revealed common cortical networks subserving participants’ aesthetic judgments. Discussions of the functions of these cortical networks suggest that aesthetic judgments within a medium are realized by a unique set of psychological processes.
    • one can argue against Dickie that our use of aesthetic concepts in ordinary language draws on, reflects, or perhaps even itself constitutes an underlying tacit theory about the nature of art and the character of aesthetic experience; our systematic use of aesthetic terms in ordinary language should yield testable predictions about the nature of art and the character of aesthetic experience. Therefore, even if, as Dickie suggests, the philosopher of art is concerned with language games, a language game can be studied empirically.

In below, the essay explores significant interactions between philosophy and psychology with special regard to the arts of literature, music, and visual art.

Literature

• Jenefer Robinson’s Deeper than Reason: The Emotions and their Role in Literature, Music and Art.
  • James’s approach to the emotions: it is physiological change that is essential to an emotion. When we perceive something that causes fear, for example, that perception directly causes bodily changes.
  • judgmentalism: A judgment is an assertion and an assertion entails that we believe its propositional content. Therefore, the judmentalist holds that in order for an emotion to take hold, it is necessary that we have certain beliefs: an emotional state is essentially a judgment.
  • Robinson’s rejection to the judgmentalism:
    • thought experiments (philosophically)
      • imagining cutting finger in half: emotion happens without belief
    • psychological
      • Zajonc has demonstrated that, at least in primitive episodes of emotional response such as the startle reaction, affect can obtain without being caused by any cognitive state
      • Zajonc and his colleagues have established that people form preferences for stimuli to which they had been previously exposed, but that were presented to them at rates too fast to allow conscious recognition
    • neuroscience
      • Joseph LeDoux: “Thinking, fast and slow”: “fast” activates the amygdala directly, but then that system is checked by processes operating in the frontal cortex (“slow”).

On the judgmentalist picture, the emotional state begins with a conscious judgment—an evaluation or appraisal of the stimuli: this causes a feeling state in me that puts me on the lookout for more relevant information and primes my behavioral response.

For Robinson, perception -> physiological -> behavioral -> cognition kicks in at this point in the process, monitoring the earlier stages of the process and either endorsing them or modifying them. Robinson believes that changes in the facial musculature and the corresponding alterations of the autonomic nervous system provide some grounds for discriminating between the so-called basic emotions (e.g., anger, disgust, fear, joy, sadness, and surprise). However, she also maintains, as do the judgmentalists, that it is primarily the cognitive processing that attends the emotion that fixes the identity of the affective state.

However, some emotions look like they require a lot of cognition in the form of some antecedent deliberation. Absorbing literature emotionally seems more like the sort of affect that is born of reflection than like a non-cognitive reflex such as the startle response.

Reading a fiction prompts our memory system to recall event types like the ones in the story that provoke immediate, non-cognitive, affective appraisals, and thereby activate physiological changes and behavioral tendencies; this then is cognitively monitored, leading to an adjustment of our appraisal of the situation, refining our understanding of it.

• cognitively monitoring our emotional responses to literature can provide the foundation of our interpretive response

=> Therefore, psychology is anything but irrelevant to the philosophy of art: it both contributes to our understanding of and enhances our capacity for narrative appreciation.

Music

• Plato’s Republic: music in different modes affects our mental states in different ways
  • music is something we make direct physical contact with; the emotional and aesthetic effects of music are directly caused by the physical properties of musical sounds.
• Eduard Hanslick’s The Beautiful in Music.
  • our affective responses to music could not involve genuine emotions, for an emotional response is not possible unless its object has some semantic content.
    • similar to judgementalist
    • in order for us to be saddened by music, the music would have to depict some particular thing that we recognize and react to in a canonical way. But music does not have the semantic resources of a language.
• Similarly, Peter Kivy
  • The inability of music to express emotion literally is a conceptual claim for Kivy—it ultimately rests on our definition of the nature of emotion and the nature of music. Philosophers would do well to recognize that although questions about the nature of music or emotions may be partly conceptual questions, they are also empirical questions. Indeed, even where they are conceptual questions, our concepts continue to evolve as they are influenced by scientific developments.
    • the study of subject and object: perceptual psychology, the science of acoustics, affective neuroscience, cognition, emotion

Music Imagery

• Wittgenstein, Ryle, and Scruton: imagining is simply an attitude we take. Rather than believing that we hear a melody when there is no actual sound present, we merely imagine the sound.

• “imagination” is currently used to refer to any one of several related concepts:

  • imagination as the production of mental imagery
  • imagination as a faculty that organizes percepts for coherent cognition (the constructive imagination)
    • Kant argues that the imagination is responsible for shaping sensory information from perception so that it is available to our understanding, or coherent to us.
  • imagination in the sense of being imaginative or creative
  • imagination as a propositional attitude taken towards fictional or imaginary propositions.

• Hume’s A Treatise of Human Nature: the imagination is a storehouse of ideas derived from past perceptions => Cognition is a capacity that always needs the imagination, since it involves the comparison of present perceptions to past perceptions, or the application of general concepts, which are formed by the imagination. Having an image of something not present also involves the imagination, since this involves either the recall of a past idea from the imagination, or the construction of a new idea out of past ideas. To be imaginative in the sense of being creative, then, is to be especially innovative in constructing new ideas, whether these ideas are imagistic (i.e., sensory) or abstract.

• Visual imagery has dominated mental imagery research in the late 20th century. These studies have demonstrated that imagery preserves key metrical properties of perceptual representations in a target modality, that it draws on the same neurophysiological processes as perception in a target modality, and that the capacity for imagery is disrupted by both focal damage and the application of repetitive transcranial magnetic stimulation (rTMS) to the discrete brain regions that realize these perceptual processes.

  • Kosslyn, Ball, and Reiser: differences in the time it takes to answer questions about the presence of targets on a memorized map are proportional to the distances one would have to scan to locate them on the original.
  • Shepard and Metzler: differences in the time it takes for participants to evaluate whether an object is identical to a target in mental rotation studies are proportional to the distance the former would have to be rotated to match the latter.
  • imaging studies demonstrate that visual imagery tasks employ the same areas of the cerebral cortex that are involved in ordinary perception
  • Bisiach and Luzzatti: selective damage to one side of the cerebral cortex causes patients to fail to perceive objects in associated areas of the visual field, generalize to visual imagery.
  • Kosslyn and his colleagues have demonstrated that the use of rTMS to disrupt activity in areas of the cerebral cortex involved in the encoding of sensory information in visual experience also disrupts the capacity in normal perceivers to use imagery and visualization to make spatial comparisons among the parts of recollected visual stimuli

• Auditory imagery experiments confirm that the metric properties (pitch, rhythm, and timbre) of auditory perception are preserved in auditory imagery: auditory imagery draws on the same neurophysiological processes in the auditory cortex as ordinary hearing, and that focal damage to these areas disrupts auditory imagery.

  • Halpern: asked participants to identify whether a lyric was part of a song or not. Those asked to mentally play through the song took longer to respond than those who were not, and differences in their response times were proportional to the musical time between the target lyrics
  • In a second experiment, Halpern asked participants to identify whether a note corresponding to a lyric was higher or lower in pitch than the starting note. Some subjects could not do this at all. However, those with musical training were better than non-musicians, and differences in response times were more pronounced than in the previous experiment.
  • the evidence seems to point towards a representation that
    • codes extension in time
    • unfolds in real time
    • has strong links between adjacent element
    • unidirectionally ordered

• Overall, these studies in psychology and neuroscience support the hypothesis that “parts of the cortex specialized for processing actual sound are also recruited to process imagined sound.” This, in turn, lends support for the phenomenological intuition that the experience of imagery is importantly like the experience of perception.

• Auditory memories seem to be stored as past auditory percepts, and can be recalled in a way that reproduces the pitch and temporal details of the original auditory experience.

  • Zatorre and Halpern review evidence that suggests that it is plausible to think that composers who report “hearing” their compositions internally are drawing on the same capacity for auditory imagery used to rehearse a remembered piece of music. Hearing a piece that one already knows involves reviving an auditory image of that piece, drawing upon memory stores that function as a kind of constructive imagination.

• neural underpinnings of auditory imagery elucidates an important component of the musical listening experience

  • musical capacities are highly variable: both listening to and performing music seems to alter the very structure of the brain - arguably, what is happening is the formation of aural/musical categories for the components of music as well as for entire compositions.
    • 2 possibilities:
      • musical ability relevant to music listening can be developed later in life
      • there are limits to what listening capacities many people can develop

Visual art

Cogintive Science: is the study of the way organisms acquire, recognize, manipulate, and use information in the production of behavior (the input and the output and their relationship): perceptual, affective, and cognitive

Current research in the cognitive neuroscience of visual art lies at the confluence of two broad research strategies:

• empirical aesthetics: examine behavioral responses to artworks.
• aesthetic experimentalism: the view that visual artists develop formal and compositional vocabularies by trial and error, or through a systematic exploration of the perceptual effects of different sets of marks, color schemes, and compositional arrangements. Therefore, we can learn about the operations of perceptual systems by examining the productive strategies of artists

cognitive neuroscience of visual art rests on the assumption that correlations between the formal features of artworks and basic neurophysiological mechanisms in the visual system can be used to explain how visual artworks work as perceptual stimuli. A good deal of research in the cognitive neuroscience of visual art is dedicated to searching out and explaining these relationships between the formal strategies employed by artists and the operations of the visual system.

• visual system is a set of evolved biological mechanisms whose function is to select information from sensory inputs that is sufficient for visual recognition and action
• The input to the visual system is replete with information about the local environment. However, only a small fraction of this information is diagnostic for the identity of an object, action, or event at any given time.
• Artists’ formal methods and vocabularies are tools for selecting features sufficient for object, action, and event recognition from ordinary perceptual experience and rendering them in a medium

Artworks can be thought of as artifacts intentionally designed to direct attention (perceptually sailent) to their artistically salient features (perceptual features responsible for the aesthetic effects and semantic associations)

3 case studies

Georges Seurat’s Bathers at Asnières

Georges Seurat used a formal technique called irradiation to enhance edges and amplify figure–ground segregation in Bathers at Asnières

• Irradiation is derived from the observation of Mach bands in ordinary perception, but has been used by painters long ago.
  • Mach band: Lateral inhibition helps boost the perception of object edges
    • A ganglion cell in the retina receives excitatory signals from photoreceptors and inhibitory signals from other ganglion cells that surround it. Ganglion cells along the bright side of the border between these regions receive less inhibition from their neighbors on the dark side. The overall response of these cells will therefore be higher than their neighbors

one function of lateral inhibition is to amplify the intensity of a feature of the sensory input that defines the boundaries between objects in perception. This in turn contributes to form recognition by enhancing the contrast between figure and ground, thereby sharpening the perception of depth in the picture plane.

An artists’ formal vocabulary => its utility in paintings is explained by neurophysiological mechanisms in the visual system.

However, although this case study demonstrates that cognitive neuroscience can contribute to our understanding of the way paintings work as perceptual stimuli, it does not establish irradiated edges as artistically salient features of Bathers at Asnières, because Mach band is everywhere in the real life but we do not find all of them aesthetically pleasing => if the nature of art can be explained in terms of ordinary psychological processes subserving our engagement with artworks, one needs an additional explanation to determine what, if anything, differentiates artworks from ordinary artifacts.

Mona Lisa

Sfumato is a formal technique in which artists blur the sharp edges that define object features in a painting. Leonardo used this technique to render the critical facial features that define Mona Lisa’s expression, the corners of her mouth and eyes. Gombrich argues that these sfumato contours introduce a degree of ambiguity into the painting that forces viewers to use their imagination to interpret an expression that cannot ever be discretely resolved.

• Livingstone has demonstrated that differences between the spatial resolution of peripheral and foveal vision explain how sfumato works to generate the dynamics of Mona Lisa’s expression
  • We are able to discern remarkably fine-grained visual detail in the central, or foveal, region of the visual field, but nearly blind to visual features defined by coarse-grained visual information; converse is true for peripheral vision. Livingstone reports that the spatial resolution of normal vision diminishes by a factor of 10 just 7 degrees from the central.
    • This difference in spatial resolution between central, or foveal, and peripheral vision is explained by the fact that the receptive fields of peripheral retinal neurons are dramatically larger than those of their foveal counterparts: foveal neurons are sensitive to sharp, narrow-luminance boundaries that carry high-spatial-frequency information, but are unable to register coarse, broad-luminance gradients, like contours rendered in sfumato, that carry low- and medium-spatial-frequency information.

Smile is “represented” using sfumato in low and medium spatial frequencies: when one foveates on Mona Lisa’s smile it disappears; the smile reappears in a viewer’s peripheral field when he or she looks away.

Gombrich VS Livingstone:

• Gombrich: a viewer’s interpretation of Mona Lisa’s expression should vary with his or her state of mind
• Livingstone: he depicted expression varies systematically with the distance of one’s focus of attention from the center of Mona Lisa’s face

However, similarly, it does not establish sfumatoed smile as artistically salient feature because of the same reason. Nonetheless, Livingstone’s discussion contributes evidence in support of a particular hypothesis about the aesthetic quality of the painting.

Theories about the nature of art and aesthetic experience rest ultimately on their consistency with facts about the way we engage with particular artworks => vision science, or psychology in general can serve as constraints to help decide among competing theories.

Slave Market with the Disappearing Bust of Voltaire

Dalí’s painting is a bistable ambiguous image (like the duck/rabbit image). The bistability is an artistically salient feature of the painting that generates both the perceptual ambiguity and the range of semantic associations that are constitutive of its identity as an artwork.

Bonnar and her colleagues used a technique called frequency-specific adaptation to demonstrate that this high- and low-spatial-frequency information was selectively used to drive the perception of either the nuns or the bust of Voltaire in the painting.

• the nuns were more apparent in the high-spatial-frequency images and the bust of Voltaire was more apparent in the low-spatial-frequency images.
• hypothesis: Frequency-specific adaptation can be used to desensitize the visual system to the particular spatial-frequency information presented in a stimulus. participants adapted to high-spatial-frequency information should see the bust of Voltaire and participants adapted to low-spatial-frequency information should see the two nuns.
• Participants were primed with either high- (first group) or low-spatial-frequency (2nd group) dynamic noise. THe first group reported seeing Voltaire; the 2nd saw nuns.

Summary

We can come to understand the nature of art and aesthetic experience by tracking the relationship between the formal and compositional structure of particular artworks and viewers’ behavioral responses.

The model that emerges defines artworks as attentional strategies intentionally designed to direct attention and enhance the perception of features diagnostic for their artistically salient aesthetic (e.g., Mona Lisa’s elusive smile) and semantic (e.g., the bistable ambiguous content of Dalí’s painting) content.

These case studies each demonstrate a role for exogenous, stimulus-driven attentional processes in our engagement with artworks. However, attention is also endogenously driven by knowledge and expectations. Research suggests that artworks are attentional strategies that harness both basic formal features and viewers’ semantic knowledge, including art historical knowledge of the unique sets of formal strategies used to categorize artworks as belonging to artistic types (e.g., the works of particular artists, schools, or epochs), to convey their content.

Challenges

• more work needs to be done to determine how to isolate and model the types of aesthetic and semantic effects used to categorize artifacts as artworks.
• one needs a way to constrain our understanding of formal structure to just those features and relations that contribute to the function of an artifact as an artwork.
  • functional rather than descriptive
  • the functional sense provides a principled means not only to recover the content of the work without demanding explicit prior knowledge of an artist’s actual intentions, but also to evaluate the adequacy of competing interpretations of those intentions.