Aphantasia: A Condition Beyond Visualization

Consider this: what if you could not visualize in your mind? It’s a perplexing notion, especially for those of us who can easily conjure mental images. Yet, this is the reality for approximately 3.9% of the population.

“Aphantasia” refers to the inability to voluntarily create mental imagery. Individuals with aphantasia cannot visualize objects in their mind's eye. For instance, if you asked them to imagine a red apple, they would struggle to form a mental picture. They understand what a red apple is but cannot visualize it.

Is this a disadvantage? While many individuals with aphantasia have thrived (I will soon mention a Disney animator who has this condition), it does present challenges that can impact visual-spatial working memory, autobiographical memory, and insight. This indicates that AI-assisted visual representation could be beneficial.

From Aphantasia to "Fantasia"

Walt Disney famously stated, “Animation can explain whatever the mind of man can conceive.” It may come as a surprise that Glen Keane, the lead animator known for characters like Ariel and the Beast, is celebrated as "one of the greatest animators in the history of hand-drawn animation," yet he lacks a visual imagination.

How Glen Keane Creates Art Without Visualization

It is essential to differentiate between the inability to evoke mental imagery and the knowledge of how an object or scene should appear. When artists with aphantasia create, they refine their work in response to external references. They possess implicit knowledge that guides them without necessarily visualizing it.

For instance, Glen Keane does not start with a blank canvas; he begins with random scribbles (which we will later compare to the "Seed" or static employed in AI-generated art) and then meticulously removes what doesn’t belong while emphasizing what does, until the desired form emerges from the chaos.

Even without visualizing a red apple, one can comprehend its structure and texture. It serves as an interesting mental challenge for those without aphantasia to try recalling an object without conjuring an image (akin to the classic dilemma: “Don’t think of an elephant.”)

Refining Creativity: The Iterative Nature of AI Art

Creating art with aphantasia is an iterative process: the artist may compare their work to their understanding or expectations of the object. This comparison need not be visual; it can involve a sense that a line feels "off" or "right." Some individuals with aphantasia indicate that they rely more on a "feeling" than a clear mental image. They may possess a sense of proportion or an aesthetic that guides their hand, even if they cannot "foresee" the final product.

Perhaps aphantasia can even provide an advantage when approached intentionally, allowing for a critical separation between an artist's intent and the final creation. Importantly, Keane does not attempt to depict what exists in his mind (and subsequently fails to match it, as what we create can never align perfectly with our “ideal vision”—the Platonic idea that all art is but imperfect copies of a perfect form). Instead, he works through the process on the page, each moment shaping the evolving piece.

The artist essentially uses the canvas as an external cognitive tool. Rather than struggling with an internal image, they engage with a concrete representation that can be analyzed, modified, or discarded.

“Look at This Stuff, Isn’t It Neat?”

I would never claim that Keane could be “replaced” or “superseded” by an AI art generator envisioning the preliminary stages of animation (although Disney has always embraced technology, even when it disrupts creative fields. For instance, Disney was among the first to adopt Xerox machines, which transformed the traditional Ink and Paint department, enabling the creation of “101 Dalmatians,” a film that would have been impossible without photocopying).

There may even be a paradoxical benefit to having aphantasia. Many Pixar artists, according to a survey conducted by co-founder Ed Catmull of 540 colleagues, possess what he describes as a “blind mind’s eye” (as opposed to their more visually-oriented production managers). This growing awareness has prompted numerous creatives to identify themselves as aphantasic.

Moreover, the externalized, iterative process of working with aphantasia can yield highly original artworks, as they are not confined to a preconceived internal image, memory, or cultural expectation. The art becomes a dynamic, evolving entity, continuously refining its own trajectory.

Fixing Flounder? AI Art in the Live-Action “The Little Mermaid”

The potential exists for more artists who feel uncreative because they cannot visualize—or have been led to believe they aren’t creative—to flourish through the use of AI visualization.

Why Promote AI Art When Aphantasic Artists Thrive?

It’s evident that individuals with aphantasia can be successful artists. There is even an exhibition at the Centre for the Study of Perceptual Experience celebrating the works of aphantasic artists. History is replete with artistic figures who have either overcome or leveraged their unique neurological traits to achieve greatness. However, quantifying how many others have been discouraged is impossible. With AI art generators, those with aphantasia gain an additional tool for exploration, creation, and self-expression.

The mission of this article is to inspire artists. If you identify with the description of aphantasia and wish to create visually, I hope you’ll dismiss traditional assumptions that artists must work from a "vision of genius" and explore AI art as a means to bring your ideas to the canvas.

If you wish to articulate your artistic concept in meaningful ways and observe its development, who can say that isn’t creative? Let’s delve into the compatibility of AI art with the aphantasic creative process.

The Relationship Between Aphantasia and AI Diffusion

In a previous article, I discussed how AI image generators operate. While much has changed since then, the fundamental principles remain intact. I want to emphasize that AI does not simply conjure images; there’s no "imagination" pulling pictures from thin air. Instead, it functions more like an image "restorer."

Diffusion is crucial in clarifying that AI art generators do not “steal” artists’ work; a common misconception is that AI art is akin to a collage, rummaging through a collection of images to locate visuals that fit the prompt, and then cutting and pasting them together into a composite image.

This notion is far from accurate. Picture diffusion as dropping ink into water; it disperses, becoming chaotic and random. An AI diffusion model, conversely, does the reverse: it takes this “dispersed” randomness and reconstructs it into a clear image.

From Diffusion Confusion to Clear Design

Thus, AI begins with a chaotic, meaningless image—essentially a pattern of pure Gaussian noise represented by a Seed number—and works to restore this ‘damaged’ image according to what it understands about how the subject should appear. It accomplishes this by training on a dataset of real images and learning how to reverse the process of converting an image into noise. Over numerous iterations, the noise is systematically refined to align with the prompt (which is unrelated to the Seed; the image’s appearance cannot be directly inferred from the seed value. The seed merely serves as a starting point for the process).

The diffusion process in AI art starts from a canvas of pure noise—like a piece of paper covered entirely in random doodles. There’s no recognizable shape or form, just chaos. Sound familiar? It resembles Glen Keane’s initial scribbles.

The AI cannot internally envision what a cat, tree, or face should look like. Yet, through iterative feedback, it can produce convincing representations of these subjects. AI transforms random noise patterns into coherent, structured images. It doesn’t “cut and paste”; it “shapes and refines.”

How AI Creates Images Without Visual Imagination

Here’s a breakdown of how Generative Adversarial Networks (GANs) operate and their similarity to aphantasia:

1. Two Neural Networks: GANs comprise a generator and a discriminator. The generator produces images, while the discriminator evaluates them.
2. Generator’s Role: The generator lacks an "internal visualization" of what an image should resemble. Instead, it generates a random Seed.
3. Discriminator’s Role: The discriminator assesses the image created by the generator and compares it to its training dataset, providing feedback about how accurate or inaccurate the generated images appear.
4. Iterative Feedback Loop: The generator and discriminator compete. Through numerous iterations and feedback, the generator enhances its output.
5. Final Outcome: After many iterations, a well-trained GAN can produce images that are nearly indistinguishable from real images to human observers.

Reinforcement Learning from Human Feedback (RLHF) is also a critical element, wherein individuals evaluate output and reward the model.

This process illustrates the exterior-focused, feedback-driven nature of both human and machine learning in generating images without true visualization. But how might the relationship between aphantasia and AI art be utilized in the artistic process to bring the artificial to life?

The Real Thing: Utilizing AI Art Authentically

The way AI generates images aligns seamlessly with how aphantasic artists develop their work. AI art can be integrated into the aphantasic creative process, ensuring that they are not faced with a blank canvas—because AI imagery should serve as a canvas, not necessarily the final product. This raises an important question about whether individuals who utilize AI art are indeed artists or possess any creative ownership over the generated images. There exists a process of refinement akin to sculpting; one does not critique the sculptor for not creating the marble.

How AI Art Could Support Aphantasic Artists

• No More "Blank Canvas": Individuals with aphantasia can provide the AI with detailed descriptions, colors, shapes, or emotions they wish to capture, allowing the AI to generate images based on these inputs.
• The Iterative Process: The capacity for rapid iteration with AI facilitates experimentation with visual concepts. Each iteration helps the creator move closer to a visual representation that resonates with their intent.
• Collaborative Creation: Aphantasic artists may find it challenging to begin with a visual concept. With AI, they can share their thoughts, regardless of how abstract or verbal, and see how the AI interprets them. The artist can then refine or discard these suggestions, shaping their vision externally with AI’s assistance.
• Expanding the Feedback Loop: Incorporating AI into the creative process broadens the traditional feedback loop. Rather than relying solely on personal intuition or external critique, aphantasic artists can utilize the AI’s interpretation of their input as a form of objective feedback.
• Tangible Feedback: Externalizing the feedback loop provides immediate, concrete responses. Instead of grappling with an ephemeral internal image, the artist engages with a physical representation.
• Mirror, Mirror: The external artwork reflects the artist’s words as images, acting as a mirror to their internal state. It may unveil aspects of the artist’s subconscious, emotions, or aesthetic preferences that they were previously unaware of.
• Diverse Aesthetic Possibilities: AI art generators can expose any artist to styles they may not have previously considered, pushing them beyond their comfort zones and leading to innovative outcomes.
• Liberation: By not being constrained by internal visualization, artists can experience both the freedom of limitless possibilities and the undefined.
• Empowerment: AI allows artists to explore visual ideas that may have once seemed unattainable due to their inability to visualize, bringing the unseen and unimagined into clear focus.

For artists facing aphantasia, AI art generators could bridge the visualization gap. For those willing to embrace AI, it can facilitate the exploration of visual concepts that previously felt out of reach.

Seeing is Believing: AI Art in Practice

I have witnessed this firsthand. I’ve guided an AI art student in her 80s who finally managed to bring a character she had "thought about for years" to life on canvas—witnessing that was a privilege. Recently, a reader shared how her late mother could have utilized AI art as a creative outlet when her rheumatism limited her artistic pursuits—though this wasn’t a case of aphantasia, it highlights another beneficial application.

And It’s Not Just for Artists

AI imagery can also assist those who are not artists but struggle with visualization. I’m an avid reader (I used to teach English literature), and a friend of mine once asked how she could "get into reading." Initially, I wondered if she faced dyslexia. However, it turned out she could read fluently but never "saw" what was described on the page.

Everything clicked: her issue wasn’t with reading but with visualization. I suggested she try reading books that had been adapted into films, watching the movie first to create mental imagery. This allowed her to hold onto the memories of what she had seen as a reference while reading, even if those memories conflicted with the book's descriptions.

By comparing and contrasting what she had already seen, she was able to leverage that knowledge to enjoy the book; she now reads constantly, particularly adaptations by Netflix. Imagine if e-readers could provide real-time tailored illustrations as one reads. (Any developers or entrepreneurs interested in collaborating on this?)

Potential Applications of AI for Aphantasia

Here are additional practical uses for the lesser-known but prevalent phenomenon of aphantasia. AI art could be beneficial in several ways:

• Externalized Visualization: For those struggling to form internal visual images, AI can serve as an external visualization tool. If someone with aphantasia wanted to recall a specific scene, they could describe it to the AI, which would generate a visual representation.
• Educational Aids: In educational environments, AI-generated visuals can assist students with aphantasia in grasping concepts often taught with a reliance on visual imagination. For instance, understanding complex spatial concepts or historical events can be enhanced with visuals tailored to individual student descriptions or inquiries.
• Therapeutic Applications: Engaging mindfully with AI art may provide therapeutic benefits. It could aid individuals with aphantasia in expressing feelings, ideas, or memories through visual mediums, fostering personal reflection.
• Enhancing Memory: By transforming memories into visual representations via AI, individuals with aphantasia can obtain tangible "snapshots" of moments or ideas, facilitating recall or sharing memories with others. Through description, refinement, and adjustment of inputs, they can arrive at a visual representation that resonates with their recollections.
• Facilitating Communication: For those with aphantasia trying to communicate a visual concept (such as describing a dream or design idea), AI can assist in translating their verbal descriptions into more visual formats.
• Interactive Narratives: Storytelling and narrative exploration can be enriched with AI visuals. Instead of merely listening to or reading a story, individuals with aphantasia can use AI tools to generate accompanying visuals, enhancing immersion. I’ve created a series of Custom GPTs called “Chat Your Own Adventure” that do precisely this. Three books are available so far; I encourage you to check them out!

Building an AI-Powered Choose Your Own Adventure Novel

Currently, I offer a Gothic fantasy titled "The Castle of Shadows," a time-sensitive tale called "Save The Titanic," and my favorite, a visual detective story set in a phantasmagorical living art world entitled "Framed for Murder."

"Framed For Murder" is the most visually-oriented; the other two feature spontaneously generated images, but readers can also prompt them to self-illustrate. All images stem from the reader’s choices in the interactive narrative.

I love hearing about your adventures in my interactive books, so if you decide to play one, please feel free to share your results and images, especially on Medium. Tag me; I’m excited to read about your experiences in the worlds I’ve created.

While AI art tools cannot replicate internal visual imagination, they offer a valuable external platform for exploration, expression, and communication. By harnessing these tools, individuals with aphantasia can navigate a world that heavily emphasizes visual concepts.

A Manifesto on the State of AI Art

AI art often faces criticism. However, it’s not merely a shortcut; it can serve as an active collaborative partner in the creative process, igniting a synthesis of ideas and visuals while acting as a catalyst for creative endeavors. It resembles having a muse that translates the abstract whispers of imagination into vivid, tangible forms, expanding the horizons of artistic expression and opening new avenues for creative exploration. AI brings the unseen and unimagined into clear focus, encouraging artists to venture into new realms of creativity.

Whether you can visualize in your mind or see only darkness, art is about self-expression. It captures feelings, ideas, or stories. By utilizing AI, anyone can align their unique perception of the world with the art they wish to create. Although there’s still work to be done in establishing fair agreements with artists whose work has contributed to AI training, we must support these efforts, advocating for a future where everyone can engage in the joy and fulfillment of visual connection.

Who is Jim the AI Whisperer?

Jim the AI Whisperer offers personalized coaching on crafting original and compelling content, as well as how to utilize AI generators to create stunning visuals. If you’re interested in learning more, feel free to reach out.

He is also available for podcasts, interviews, fine-tuning AI prompts, and developing prompt libraries and professional AI artistry for businesses.

You might enjoy these related articles from Jim the AI Whisperer:

• All My AI Art Think-pieces and Editorials: Perspectives on AI art and culture.
• See All My AI Art "How to" Guides and Tips: Discover innovative ways to use AI art generators.