One claim I often find myself resisting is the familiar maxim, “learning to draw is learning to see.” The phrase has considerable cachet, and once upon a time, I found myself using it in a loose, colloquial sense. Over time, however, I stopped using it—and now go so far as to address it when it is brought up in discussions about education—because it tends to encourage the mistaken expectation that training in drawing makes vision itself more “accurate.” Drawing practice can certainly refine attention, discrimination, and the coordination of perception with mark-making, but it does not transform the visual system into an objective, or even more objective, measuring instrument.
My objection is not that such training produces no meaningful perceptual change; it plainly can. My concern is with what the word accurately tends to imply in this context. Human vision is not a transparent window through which the objective properties of the physical world are simply presented to consciousness. This has been understood in some form for centuries. As George Berkeley argued in 1709, the physical sources of visual stimulation are not directly available in experience. Modern vision science provides the biological basis for that conclusion. The visual system does not measure the objective properties of the world. It registers patterns of light at the retina, but those patterns do not uniquely specify the physical conditions that produced them. Reflectance, illumination, transmittance, distance, size, orientation, and other variables are conflated in the retinal stimulus. There is therefore no stage at which the nervous system directly samples or recovers the world’s objective properties as measurements. What we see is a biologically generated perceptual response to inherently ambiguous stimulation, shaped by the accumulated relation between retinal patterns and successful behavior. In this sense, visual perception is not veridical: percepts are not measurements or reconstructions of physical reality, but behaviorally useful products of the visual system.
Consequently, observational drawing does not teach a person to bypass perception or gain direct access to the world as it exists independently of an observer. What it can develop is a specialized repertoire of attentional, perceptual, judgmental, and motor habits. Through sustained practice, the artist becomes increasingly responsive to particular relationships within visual experience—relative angles, proportions, intervals, alignments, value differences, contours, negative spaces, and spatial relations—that are useful for producing a desired representation.
The important word here is relative. A trained artist may become better at judging that one perceived edge is steeper than another, that one interval is shorter than another, or that two regions produce more similar value experiences than their contexts initially suggest. The artist may also become better at suppressing familiar conceptual schemas—for example, the tendency to draw what an object is known to be rather than responding to how it appears under a particular set of viewing conditions. These are genuine and consequential achievements. They can produce drawings that conform more closely to a specified representational criterion, such as measured proportions, a chosen projective geometry, a photographic projection, or observers’ judgments of resemblance.
This is not equivalent to acquiring an objective form of sight. Believe it or not, this possibility has been tested directly. Perdreau and Cavanagh asked whether trained artists possess a more veridical form of perception—that is, whether artistic training grants greater access to lower-level visual information before ordinary perceptual interpretation and constancy mechanisms take effect. Their results did not support the idea that artists possess a generally more objective visual experience. Artistic expertise may improve the selection and use of information relevant to depiction, but it does not appear to reveal an unprocessed or physically faithful visual world. Perceptual training in the realm of observational visual art is better understood as the cultivation of task-specific perceptual–motor expertise. Practice establishes and reinforces increasingly reliable relations among perceptual experiences, comparative judgments, and representational actions: attending to a relation, making a mark, comparing the resulting mark with the current perceptual experience, detecting a discrepancy, and revising the mark. With practice, this perception–action cycle can become faster, more selective, more stable, and more automatic.
Written language offers a useful analogy. The printed letter “K” does not intrinsically contain the spoken sound associated with it. The visible character and the speech sound are physically different events. Through learning, however, a strong association is established between them, so that encountering the character can automatically evoke its name or sound. The sound can, of course, be acoustically measured; the relevant point is that the relation between the visible mark and the experienced sound is learned rather than physically contained in the character.
A similar process occurs in drawing. What an artist experiences as a line may arise from a perceived boundary, contour, occlusion, luminance gradient, or other discontinuity in the retinal stimulus. The artist does not receive the physical properties of the source as objective measurements. Instead, the visual system generates a context-dependent perceptual experience, and drawing practice establishes increasingly useful associations between aspects of that experience and particular representational actions. A perceived tilt becomes associated with a particular directional movement; a perceived proportion with a particular spacing of marks; a perceived value relation with a particular distribution of graphite, ink, or pigment.
Feedback allows these associations to be adjusted and stabilized. The artist compares the drawing not directly with objective reality, but with a perceptual experience, an instrumentally derived measurement, a percept surrogate, a geometrical construction, or some other selected criterion. Improvement, therefore, consists in better-regulated performance relative to that criterion. If one wishes to call this improvement “greater visual accuracy,” the term is defensible only in an operational and task-relative sense. It may refer to closer agreement with measured proportions, a specified projection, a set of geometrical constraints, or judgments of resemblance. It should not be taken to mean that the artist’s visual system has become more capable of objectively measuring the physical world.
What should be resisted, then, is the stronger interpretation that observational training transforms vision into an objective measuring system or gives the artist perceptual access to physical reality as such. Artists do not escape the biological and inferential conditions of vision. They become more skillful at coordinating attention, judgment, and action within those conditions. Thus, learning observational representationalism may teach a person to attend differently, compare more systematically, inhibit familiar conceptual habits, and coordinate perceptual judgments with mark-making more effectively. It can produce representations that satisfy chosen criteria with greater consistency and precision. That is an important form of expertise. It is not the acquisition of direct perceptual access to the objective properties of the world.
Resources:
Perdreau, F., & Cavanagh, P. (2013). Is artists’ perception more veridical? Frontiers in Neuroscience, 7, 6.
Kozbelt, A. (2017). Learning to see by learning to draw: Probing the perceptual bases and consequences of highly skilled artistic drawing. High Ability Studies.
Chamberlain, R., Drake, J. E., Kozbelt, A., et al. (2019). Artists as experts in visual cognition: An update. Psychology of Aesthetics, Creativity, and the Arts.
Ostrofsky, J., Kozbelt, A., & Seidel, A. (2012). Perceptual constancies and visual selection as predictors of realistic drawing skill. Psychology of Aesthetics, Creativity, and the Arts, 6(2), 124–136.
Seeley, W., & Kozbelt, A. (2008). Art, artists, and perception: A model for premotor contributions to perceptual analysis and form recognition. Philosophical Psychology.
Lou, L. (2018). Artists’ innocent eye as extended proximal mode of vision. Art & Perception, 6(1), 1–36.
Kozbelt, A., & Kantrowitz, A. (2019). Talent and ability in drawing and visual art. In The Oxford Handbook of Expertise.
Purves, D., Morgenstern, Y., & Wojtach, W. T. (2015). Perception and reality: Why a wholly empirical paradigm is needed to understand vision. Frontiers in Systems Neuroscience, 9, 156.
Marr, D. (2010). Vision: A Computational Investigation into the Human Representation and Processing of Visual Information. MIT Press.
Palmer, S. E. (1999). Vision Science: Photons to Phenomenology. MIT Press.
Berkeley, G. (1709). An Essay Towards a New Theory of Vision.
)
