A key feature of Bullot and Reber's (B&R's) psycho-historical framework is that art appreciation is influenced both by basic perceptual processing of visual information and the appreciator's knowledge of the historical context in which the artwork emerged. We applaud this interdisciplinary effort. Here, we further suggest that this aspect of the psycho-historical framework can be profitably extended to understanding artistic production, by facilitating a parallel synthesis of historically independent (or even antagonistic) perspectives on that issue.
Consider realistic observational drawing, a complex skill that involves visually perceiving an object or scene and manually transcribing aspects of the object or scene onto a surface, with the goal of creating a recognizable depiction. Over the last several decades, researchers have explored the psychological factors undergirding individual variability in realistic drawing ability with increasing intensity and sophistication. B&R's perception/knowledge distinction echoes how different camps of researchers have approached individual differences in representational drawing ability, with some emphasizing basic perceptual processing and others emphasizing higher-order cognition and the knowledge-driven selection of relevant visual information.
Empirical research on the perception side is intellectually descended from the so-called “innocent eye” hypothesis, whereby the perceptual constancies that facilitate everyday perception act to inhibit accurate perception of viewpoint-dependent or transient visual information that is essential for accurate rendering. In this view, drawing skill is regarded as mainly the result of a capacity to suppress such perceptual constancies in order to accurately perceive the to-be-drawn stimulus.
Despite some empirical support for the importance of early perception and the deleterious impact of strong shape or size constancy on accurate drawing (e.g., Cohen & Jones Reference Cohen and Jones2008; Mitchell et al. Reference Mitchell, Ropar, Ackroyd and Rajendran2005), there are reasons to believe that this explanation is incomplete. First, it mainly addresses why most people are poor at drawing, rather than how artists excel at it. Second, it fails to address logical inconsistencies of purely bottom-up perceptual mechanisms that are related to the inverse problem in vision (Gombrich Reference Gombrich1960). Third, while empirical evidence shows that non-artists are less able to overcome perceptual constancies than trained artists, even artists can only partially suppress them; indeed, artists show far greater size estimation errors (and hence size constancy effects) as a result of depth cues in a display, compared to non-artists' baseline error performance in a display lacking depth cues (Ostrofsky et al. Reference Ostrofsky, Kozbelt and Seidel2012).
In contrast to the predominantly perception-based account of drawing skill, our lab has focused on the role of the selection of relevant visual information driven by artists' domain-specific knowledge, which is acquired through intensive training. We regard artists' advantages in drawing as stemming from multiple stages of visual attention, rather than mainly from early perception (Ostrofsky & Kozbelt Reference Ostrofsky, Kozbelt, Kantrowitz, Brew and Fava2011). Specifically, artists' domain-specific knowledge of the structure of common objects and methods of depiction in particular artistic media functions to harness and selectively focus attention on aspects of objects that are crucial for recognition, while deemphasizing less important visual information. We have tested the importance of selection using tasks where participants can use only a small number of line segments to depict a complex object like a face, finding that artists produce renderings that are judged as more accurate, compared to those created by non-artists (Kozbelt et al. Reference Kozbelt, Seidel, ElBassiouny, Mark and Owen2010). This result reinforces the importance of artists' domain-specific knowledge and higher-order attentional and cognitive processes for understanding skilled drawing – in contrast to accounts emphasizing only early stages of perception, in which knowledge is regarded mainly as interfering with performance.
This integrative perspective is strongly influenced by art historian E.H. Gombrich, a significant figure in B&R's target article, and in our view a scholar whose work represents a prescient and exemplary manifestation of the psycho-historical approach. This is particularly true in his application of psychological schemata for understanding many developments within art history. A Gombrich-spirited approach, informed by contemporary psychological research, can readily be assimilated into B&R's proposed framework and applied to artistic production.
Consider the problem of understanding the history of realism, one of Gombrich's own research foci. The problems facing someone wishing to create a convincing representational depiction of the visible world are legion and can only be solved incrementally by the accumulation of organized, specialized knowledge. Artistic realism has a history precisely for this reason: artists gradually acquire (and then pass on to later generations of artists) knowledge of various depictive heuristics, which serve the goal of creating increasingly more convincing renderings.
We believe that Gombrich's account of realism makes significant contact with B&R's psycho-historical model, because artists' technical knowledge itself constitutes a major component (if not the only component) of the historical context of the emergence of artworks. Because artists' schemata evolve over time, to understand how a particular image was created, historical knowledge of the methods of artistic production, together with the depictive possibilities of different artistic media (e.g., dry-point etching can achieve effects that thick charcoal cannot, and vice versa), is critical. This analogy suggests that artistic production, like artistic appreciation, can only be fully understood through a strong historical lens.
However, we note that this historically informed psychological approach also has the potential to yield a new generation of universalist claims. Because artists creating realistic depictions must solve many of the same problems as the visual system generally (Kozbelt & Seeley Reference Kozbelt and Seeley2007), it is likely that many depictive heuristics developed throughout art history tap into fundamental aspects of visual processing. For example, compared to non-artists, artists spontaneously deploy more nonaccidental properties of objects, like vertices, in their renderings (Ostrofsky et al. Reference Ostrofsky, Kozbelt and Seidel2012), which aid identification of depicted objects along the lines of some cognitive theories of object recognition (e.g., Biederman Reference Biederman1987). Over the long term, a richly developed psycho-historical approach might accumulate numerous such convergences between artistic practice and psychological theory, in line with universalist proposals. This remains an open question – one that is probably best tackled by empirically integrating psychological and historical accounts in the near term, as suggested by B&R.
A key feature of Bullot and Reber's (B&R's) psycho-historical framework is that art appreciation is influenced both by basic perceptual processing of visual information and the appreciator's knowledge of the historical context in which the artwork emerged. We applaud this interdisciplinary effort. Here, we further suggest that this aspect of the psycho-historical framework can be profitably extended to understanding artistic production, by facilitating a parallel synthesis of historically independent (or even antagonistic) perspectives on that issue.
Consider realistic observational drawing, a complex skill that involves visually perceiving an object or scene and manually transcribing aspects of the object or scene onto a surface, with the goal of creating a recognizable depiction. Over the last several decades, researchers have explored the psychological factors undergirding individual variability in realistic drawing ability with increasing intensity and sophistication. B&R's perception/knowledge distinction echoes how different camps of researchers have approached individual differences in representational drawing ability, with some emphasizing basic perceptual processing and others emphasizing higher-order cognition and the knowledge-driven selection of relevant visual information.
Empirical research on the perception side is intellectually descended from the so-called “innocent eye” hypothesis, whereby the perceptual constancies that facilitate everyday perception act to inhibit accurate perception of viewpoint-dependent or transient visual information that is essential for accurate rendering. In this view, drawing skill is regarded as mainly the result of a capacity to suppress such perceptual constancies in order to accurately perceive the to-be-drawn stimulus.
Despite some empirical support for the importance of early perception and the deleterious impact of strong shape or size constancy on accurate drawing (e.g., Cohen & Jones Reference Cohen and Jones2008; Mitchell et al. Reference Mitchell, Ropar, Ackroyd and Rajendran2005), there are reasons to believe that this explanation is incomplete. First, it mainly addresses why most people are poor at drawing, rather than how artists excel at it. Second, it fails to address logical inconsistencies of purely bottom-up perceptual mechanisms that are related to the inverse problem in vision (Gombrich Reference Gombrich1960). Third, while empirical evidence shows that non-artists are less able to overcome perceptual constancies than trained artists, even artists can only partially suppress them; indeed, artists show far greater size estimation errors (and hence size constancy effects) as a result of depth cues in a display, compared to non-artists' baseline error performance in a display lacking depth cues (Ostrofsky et al. Reference Ostrofsky, Kozbelt and Seidel2012).
In contrast to the predominantly perception-based account of drawing skill, our lab has focused on the role of the selection of relevant visual information driven by artists' domain-specific knowledge, which is acquired through intensive training. We regard artists' advantages in drawing as stemming from multiple stages of visual attention, rather than mainly from early perception (Ostrofsky & Kozbelt Reference Ostrofsky, Kozbelt, Kantrowitz, Brew and Fava2011). Specifically, artists' domain-specific knowledge of the structure of common objects and methods of depiction in particular artistic media functions to harness and selectively focus attention on aspects of objects that are crucial for recognition, while deemphasizing less important visual information. We have tested the importance of selection using tasks where participants can use only a small number of line segments to depict a complex object like a face, finding that artists produce renderings that are judged as more accurate, compared to those created by non-artists (Kozbelt et al. Reference Kozbelt, Seidel, ElBassiouny, Mark and Owen2010). This result reinforces the importance of artists' domain-specific knowledge and higher-order attentional and cognitive processes for understanding skilled drawing – in contrast to accounts emphasizing only early stages of perception, in which knowledge is regarded mainly as interfering with performance.
This integrative perspective is strongly influenced by art historian E.H. Gombrich, a significant figure in B&R's target article, and in our view a scholar whose work represents a prescient and exemplary manifestation of the psycho-historical approach. This is particularly true in his application of psychological schemata for understanding many developments within art history. A Gombrich-spirited approach, informed by contemporary psychological research, can readily be assimilated into B&R's proposed framework and applied to artistic production.
Consider the problem of understanding the history of realism, one of Gombrich's own research foci. The problems facing someone wishing to create a convincing representational depiction of the visible world are legion and can only be solved incrementally by the accumulation of organized, specialized knowledge. Artistic realism has a history precisely for this reason: artists gradually acquire (and then pass on to later generations of artists) knowledge of various depictive heuristics, which serve the goal of creating increasingly more convincing renderings.
We believe that Gombrich's account of realism makes significant contact with B&R's psycho-historical model, because artists' technical knowledge itself constitutes a major component (if not the only component) of the historical context of the emergence of artworks. Because artists' schemata evolve over time, to understand how a particular image was created, historical knowledge of the methods of artistic production, together with the depictive possibilities of different artistic media (e.g., dry-point etching can achieve effects that thick charcoal cannot, and vice versa), is critical. This analogy suggests that artistic production, like artistic appreciation, can only be fully understood through a strong historical lens.
However, we note that this historically informed psychological approach also has the potential to yield a new generation of universalist claims. Because artists creating realistic depictions must solve many of the same problems as the visual system generally (Kozbelt & Seeley Reference Kozbelt and Seeley2007), it is likely that many depictive heuristics developed throughout art history tap into fundamental aspects of visual processing. For example, compared to non-artists, artists spontaneously deploy more nonaccidental properties of objects, like vertices, in their renderings (Ostrofsky et al. Reference Ostrofsky, Kozbelt and Seidel2012), which aid identification of depicted objects along the lines of some cognitive theories of object recognition (e.g., Biederman Reference Biederman1987). Over the long term, a richly developed psycho-historical approach might accumulate numerous such convergences between artistic practice and psychological theory, in line with universalist proposals. This remains an open question – one that is probably best tackled by empirically integrating psychological and historical accounts in the near term, as suggested by B&R.