Visual encoding of visual art: from features to aesthetics

Mapping the range of pigment, luminance and lightness in vision and art

Alan Gilchrist

The high dynamic range of most natural images, due to variations in both reflectance and illumination level, poses a challenge both for the visual system and for the artist. The visual system must assign a range of reflectances no greater than about 30:1 (from 90% to 3%) to a range of image luminances that often far exceeds 1000:1. The painter must represent that large range of intensities using the same (30:1) low range of reflectances, even after Van Eyck extended that range by shifting from tempera to oil. While the visual system organizes these high range images into regions of high and low illumination, each comprising a limited range (< 30:1) the artist must compress the high range image into the limited range of pigment. When the painting is viewed, the observer's visual system then expands the compressed range. We measured this range mapping in paintings and abstract laboratory patterns, both low-range and very high range (>5000:1), finding a tendency to normalize the perceived range to the white-to-black range. We demonstrated that use of a 3D canvas (Patrick Hughes-type reverspective) can extend the range by exploiting directional illumination on the canvas, producing a genuine perception of light-emitting self-luminosity.

Art appreciation: the interplay of perception, emotion and expertise

Helmut Leder

Art is a unique feature of human experience and several approaches aim to understand what the psychological aspects of this uniqueness are. Art appreciation involves the complex interplay among stimuli, perceiver and contexts, which have been discussed as eliciting a special combination of aesthetic judgments and aesthetic emotions. Based on our model of aesthetic appreciation (Leder et al., 2004), we conducted studies to understand the nature of stylistic processing (Augustin et al., 2008), the dependence of art appreciation of the class of artworks (Belke et al. 2010) as well as the complex interplay of the variables involved between these factors. Regarding the more complex interplay, we conducted a study in which we measured differences in preferences for classical, abstract, and modern artworks (Leder et al., 2011) and using structural equation modeling assessed the contribution of emotion, arousal, and comprehension and expertise as determining factors of art appreciation. The implications of these studies are discussed in respect to underlying theoretical foundations in particular regarding the interplay of perception, emotion and knowledge.

On the Sources of Visual Invariants in Artistic Composition

Christopher Tyler

Composition in painting derives from a wide array of influences. Here, the focus is on the visual invariants that govern the 2D layout within the available frame, and the question of which brain structures may be involved in the process. One well-known compositional formalism is the visual pyramid, which has been known as a key principle of compositional geometry since the time of Leonardo and Raphael. Originally centered in the frame, in later centuries the compositional pyramid could be shifted laterally to introduce asymmetries. What is less recognized is that the pyramid is not just a geometrical format, but typically peaks at the eye of a figure in the role of an observer of the action below. The same principle is found in portraits, in which one eye is usually close to the center line, while the body forms a pyramid expressing a variety of facets of the sitter's life. This 'center of consciousness' of the observing eye in the structure of many representational paintings takes the geometric concept to a new level of perceptual/inferential analysis. In terms of brain structures involved in guiding these compositional choices, the lateral occipital complex (LOC) plays a key role in encoding the long-range structure, balance and symmetry aspects of these compositional principles, but the concept of the observing eye implies a novel projection of the mirror neuron system to the inverse concept of 'being mirrored' by an external focus of observation, which evidently plays a key role in many forms of artistic representation. References Tyler, C.W. (2007) Some principles of spatial organization in art. Spatial Vision, 20, 509-530. Chen, C.C., Kao, C., Tyler, C.W. (2006) Face configuration processing in the human brain: the role of symmetry. Cereb Cortex, 7, 1423-32. Tyler, C.W. (1998) Painters centre one eye in portraits. Nature 392, 877-878.

Tracking perceptual learning in visual art students

Alexander Schlegel, Sergey Fogelson, Prescott Alexander, Xueting Li, Zhengang Lu, Peter Tse and Ming Meng

Visual artists' expertise comes in part from their ability to translate their rich and precise perception of the world into controlled actions such as drawing and painting. Graham & Meng [2011, VSS] and Perdreau & Cavanagh [2011, Frontiers in Human Neuroscience, 5] have found conflicting evidence regarding the ability of professional artists to 'see through' visual illusions that arise during mid-level visual processing. Even if visual artists cannot circumvent these processes, their ability to perceive and represent their environment is clearly exceptional. How might acquiring these abilities be reflected in the plasticity of adult brains? We investigated the effects of visual art training on behavior as well as the structure and function of the human brain by tracking undergraduate college students as they took an intensive course in painting or drawing, as well as control group students taking either organic chemistry or engineering problem-solving courses. We collected monthly high-resolution structural and DTI scans as well as functional scans while subjects a) made judgments about Craik-O'Brien-Cornsweet and Müller-Lyer illusions and b) created gesture drawings from observation of human figures. Our findings shed light on the neural plasticity that allows visual artists to develop their skillful observation and manipulation of their environment.

Representation and aesthetics of the human face in portraiture

Pamela Pallett, Daniel Graham, Helmut Leder and Ming Meng

Egyptian Queen Nefertiti is renowned for her elegant beauty, but do her portraits and busts reflect her real face? Here we examine how representations of the face in portraiture relate to human visual encoding of natural faces, and how the aesthetics of portraits relates to preferences for natural faces. In Experiment 1, participants made paired comparisons of attractiveness between different levels of composite-averaged portraits. We report that, as with natural frontal faces, portraits show norm-based coding properties with respect to preference: averaged portraits become more attractive as the number of contributing faces increases. In Experiment 2, we assessed the role of facial feature arrangement in this averageness effect. Participants compared the attractiveness of portraits with identical facial features but different eye-mouth or interocular distances. We again find that the average facial feature arrangement is preferred. However, we also find that the structure of the average portrait feature arrangement is significantly different from that of natural faces. Our results suggest a common norm-based encoding for portraits and photographs, even though portraits on average may not be faithful representations of the natural face and humans may appreciate a different form of aesthetics for portraits compared to natural faces.

Unnatural art, natural brain

Patrick Cavanagh

Artists represent the world quite realistically using a naive physics for the properties of light, shadow, and materials. In many cases, this unnatural representation appears to be correct, supporting an appropriate perception of the scene. We do not notice that this naïve physics deviates significantly from natural physics. These deviations reveal the properties of the internal physics and the visual processes that realize them. Importantly, paintings explore this naive physics in ways that that photographs and real scenes cannot. We will look at the depiction of material properties and high contrast rendering to see examples of the discoveries about the brain that are found in works of art.