What Does Realistic Look Like?

Anthony Waichulis is a celebrated Trompe L’Oeil painter from Pennsylvania, US. Since 2011, Waichulis has worked with Tim Reynolds, who co-founded Jane Street LLC and founded the Tim Reynolds Foundation, to set up Ani Art Academies, which provides generous scholarships to aspiring artists from around the world. Waichulis, who’s exhibited as far and wide as The Smithsonian Institute, Art Basel and Beijing World Art Museum, devises the Ani Art Academies’ curriculum. In this article, he answers a question often put to him by his students, ‘How do you get your paintings to look so realistic?’ Waichulis prefers to start by exploring a more fundamental problem: ‘What does “realistic” look like?’

 ‘What Does Realistic Look Like?’

I’m often asked, ‘How do you get your paintings to look so realistic?’ It’s a great question that opens the door to an enjoyable conversation about art and visual perception. I inevitably counter with a question of my own: ‘What does realistic look like?’ Many will often gravitate to the idea that a representation is ‘realistic’ if it corresponds to ‘real world’ properties. And they’d be right! The Oxford Dictionary defines ‘realistic’ as ‘[r]epresenting things in a way that is accurate and true to life.’ Merriam Webster plumps for ‘the theory or practice of fidelity in art and literature to nature or to real life and to accurate representation without idealisation; accurately representing what is natural or real; convincingly rendered to appear natural.’ Now, while these usages may often work just fine for many day-to-day situations, they can be problematic when it gets to the nitty-gritty of visual perception and/or visual art. Simply put, we don’t perceive real world properties ‘as they really are’ because visual perception is non-veridical. What we see does not accord with the physical properties of the environment or even with the retinal light patterns that appear in our eyes.  

It’s not uncommon, however, to intuitively regard the eyes as tiny cameras that present your brain with an accurate image for review. It’s easy to fall for the ‘homunculus fallacy’ whereby we imagine that someone or something sits in the brain reviewing the images presented therein. (The word ‘homunculus’ means ‘little man’.) Even though some can push past this ‘homunculus fallacy’, they still manage to get mired in the idea that vision is, in some part, objective. Unfortunately, these ideas could not be further from the truth. We’ve come to learn that the function of the visual system is not to provide a veridical window on the environment, but rather, has evolved to provide adaptive, neural responses that are useful in achieving successful behaviour amid a dynamic, and sometimes hostile, environment. As such, the idea that ‘realistic’ describes a correspondence between visual representation and real-world just doesn’t hold much water.

Anthony Waichulis’ Trifecta (2018)

That’s not to say that conceiving of visual perception via the ‘homunculus fallacy’ is stupid or wrong; it’s just to say that shortcuts such as the fallacy are problematic when they serve as the foundation for more robust concepts. Take, for e.g., the simple concept of mixing colours. Artists have long mixed various piles of paint on their palettes, but not one of them has ever physically mixed two or more colours together. You see, colour is not a property of the environment, but a perceptual experience that results from neural activity. Hence, we can never physically mix any colour with another. We can only mix certain materials, which, when viewed, elicit a perceptual response that is different from those responses elicited from each component material. Now, this might seem rather pedantic and, yes, it’s easier to think of this as just ‘mixing colours’. But, when you start delving into increasingly complex concepts about colour and perception, this semantic shortcut, ‘mixing colours’, may make less and less sense. With regard to ‘realistic’, I want my students to go beyond the colloquial and into the nitty-gritty of perception proper.

What I tell them isn’t meant as a statement on the nature of reality (for e.g., physicalism, idealism, material existence, and so on). Instead, it’s supposed to challenge long-standing assumptions about the nature of ‘realistic’ that can potentially impact our efforts as artists. As such, it’s my position that when we describe something as ‘realistic’, we’re not describing its real-world properties or even the level of kinship between our perception of those properties and the properties themselves; instead, we’re describing something that I find far more fascinating:

Realistic is the degree of relative similarity between a perceptual response to a surrogate, simulation, or other representation and the past perceptual responses to the stimulus, stimulus components, or experience that is being represented.

Now, some may be quick to point out that the idea of comparing stored experiences with new ones seems like just another problematic appeal to the brain-as-a-computer metaphor—in this case, an attempt to imply that we can ‘store’ imagery away in some neural memory register for comparison tasks. This isn’t the case. In fact, I’d go so far as to state that we can indeed experience similarity between two perceptual experiences without storing one single objectively recorded snapshot for that internal homunculus.

In a 2016 essay for Aeon titled ‘The Empty Brain’, psychologist Robert Epstein writes, ‘No matter how hard they try, brain scientists and cognitive psychologists will never find a copy of Beethoven’s 5th Symphony in the brain–or copies of words, pictures, grammatical rules or any other kinds of environmental stimuli.’ To demonstrate support for this this claim, Mr. Epstein goes on to present a fun exercise that he often presents to his students. He begins by asking a willing participant to draw a dollar bill from memory.  Afterward, the drawing is hidden away, and the participant is presented with an actual dollar bill to use as a reference source for a second drawing. When the second drawing is completed, the first is presented again for comparison. Of the results presented in the article Mr. Epstein explains, ‘As you can see, the drawing made in the absence of the dollar bill is horrible compared with the drawing made from an exemplar, even though Jinny (the participant) has seen a dollar bill thousands of times. What is the problem? Don’t we have a ‘representation’ of the dollar bill ‘stored’ in a ‘memory register’ in our brains? Can’t we just ‘retrieve’ it and use it to make our drawing? Obviously not, and a thousand years of neuroscience will never locate a representation of a dollar bill stored inside the human brain for the simple reason that it is not there to be found.’

While I would indeed agree that we do not carry an objectively recorded image of a dollar bill in some memory register in our brain—that does not mean that there is nothing that we might define or interpret as a ‘stored’ representation. In fact, if there was nothing to draw from, why would a participant’s first attempt be populated with anything?  If there was no endogenous information source, shouldn’t the first image be completely blank?  But we do find some content in those first attempts. The participants are populating the paper with something that represents, to them at least, a dollar bill.

To effectively tackle the issue of why Mr. Epstein’s initial drawing attempts do in fact contain something rather than nothing, we need to understand some basic workings of the human brain. The brain is incredibly malleable or ‘plastic’ in a great many ways. It is often busy forging new connections, pruning existing ones, strengthening those circuits that show great promise and coordination and diminishing those that are deemed a hot mess. It’s this neural maintenance plan that allows us to survive, prosper and reproduce within a highly dynamic environment.

For e.g., every time you visually encounter an apple, your brain responds with an array of neural activity. Not only does this activity determine your visual experience, but it also uses the experience to help determine your future responses to apples. With continued experience, the perceptual responses that are demonstrated to be useful during a visually guided interaction with an apple are reinforced, while others that are deemed less useful, are not. This will result in a strengthened connection between those neural circuits that have facilitated successful interactions with an apple (often across various viewing conditions) in the past. Now, if someone asks me to imagine or picture an apple, I don’t perceive an apple magically popping into existence (that would be rather terrifying and not very useful)—rather, I respond with the ‘higher-level’ cultivated neural activity that I have cognitively bound to the concept of ‘apple’. This activity is often experienced as an abstract, spatially imprecise representation with the most strengthened attributes or components ‘appearing’ most prominent. It is this mental representation that the participants in Epstein’s exercise draw from first.

Second, while the brain-as-a-computer metaphor is problematic, for sure, the argument that ‘the brain is different from a computer and therefore holds no information whatsoever’ is clearly a non-sequitur. We can absolutely view the configurations of neural circuits to be a form of information storage as it is the manner in which these circuits function that provides us with the content of our perceptions as well as the basis for our memories.

As I stated earlier, through experience, the responses that I developed from visually guided interactions with an ‘apple’ have become bound to certain higher-level neural activities (including recognition and categorisation activity we might think of here as cognitive ‘hashtags’) so as to cultivate increasingly useful and organised assignments of meaning to sensations. Therefore, even though I haven’t got that clear photograph of an apple in my head for comparisons, I can experience the activity of those higher-level cognitive hashtags during perceptual events as degrees of perceived similarity. In other words, I don’t need a stored picture of an apple in some biological memory register to experience that a current surrogate is similar to a past percept. I need only experience the similarity in neural activity between a current and past brain state.

Thus, in the context of representational art, experienced similarity between a perceptual response to a representation and the past perceptual responses to the stimulus, stimulus components, or experience that is being represented, is exactly what realistic looks like.