Visual Thinking: How to Reason and Create in Images
Visual thinking is the use of spatial and imagistic representations to reason, plan, and generate ideas. Most discussions of creative thinking focus on verbal or symbolic cognition — concepts, words, logical propositions. Visual thinking is the other half that those discussions typically omit.
In 1969, Rudolf Arnheim argued in Visual Thinking that visual perception is not a pre-cognitive input to thought but is itself a form of thinking. Seeing is not a stage before reasoning — it is reasoning conducted through spatial relationships, proportions, and configurations. This argument was controversial at publication, but the evidence for robust visual and spatial cognition as independent modes of intelligence has grown substantially in the decades since.
What Visual Thinking Actually Is
Visual thinking uses mental imagery, spatial relationships, and visual representations to process information and generate new combinations. It is not limited to people who describe themselves as "visual thinkers" or who prefer diagrams to text. Almost everyone uses visual and spatial processing, but the degree to which it contributes to problem-solving varies considerably across individuals and situations.
Temple Grandin has described three subtypes of visual thinking based on her research and personal experience: object visualization (thinking in photographic images), spatial visualization (thinking in patterns and relationships between parts), and verbal-linguistic thinking (which overlaps less with spatial modes). These aren't rigid categories — most people use all three — but they suggest that "visual thinking" covers a family of related cognitive modes rather than a single capacity.
The spatial visualization subtype is most directly relevant to creative problem-solving. This mode represents how parts relate to each other, how systems connect, and how changes in one area propagate through a structure. Engineers, architects, surgeons, and chess players rely heavily on it. So do scientists working at scales — molecular, astronomical — where direct perception is impossible and mental models must substitute.
The Cognitive Science
Roger Shepard's research on mental rotation in the 1970s established that visual imagery obeys spatial constraints. When people mentally rotate objects, their response time scales with the angle of rotation — the same relationship that holds for physically rotating objects in the world. The mind represents and manipulates spatial information using processes that mirror physical space, not just abstract propositions about it.
Barbara Tversky at Stanford has shown that spatial cognition is foundational to abstract thought. Her work demonstrates that people understand time, causality, and social relationships through spatial metaphors — earlier events are "behind" us, causes "push" effects, importance is "above." This isn't merely linguistic convention; the spatial frame actively structures the reasoning. When visual representations are removed from tasks that typically rely on them, performance degrades in ways that verbal alternatives don't fully compensate for.
This has direct implications for creative work. When you work a problem exclusively in verbal or symbolic mode, you access one portion of your cognitive architecture. Introducing spatial or visual representation — a sketch, a diagram, a physical arrangement — engages additional processing that can reveal relationships invisible in the purely verbal representation.
Visual Thinking and Creative Breakthroughs
Richard Feynman described his creative process as running mental movies — watching physical processes unfold spatially and then encoding the observations in mathematics. Einstein described the thought experiments behind special relativity as initially visual and kinesthetic: imagining riding alongside a beam of light, feeling what that would be like, before writing an equation. These accounts are anecdotal, but they're consistent with what cognitive science finds about how domain experts represent problems.
Research on insight problem-solving offers converging evidence. Many classic insight puzzles — the nine-dot problem, matchstick problems, visual paradoxes — resist purely analytical approaches and yield when the solver adopts a spatial or visual representation that makes the hidden structure apparent. The insight doesn't come from reasoning harder within the original representation; it comes from switching to a representation where the solution is visible.
Abstract thinking at its best involves moving fluidly between levels of representation — specific to general, concrete to schematic. Visual thinking is one primary route to the schematic level: diagrams strip away surface detail and make structural relationships legible in a way that dense prose often obscures.
The Diagram Advantage
There is robust evidence that external visual representations — diagrams, sketches, spatial layouts — outperform purely verbal representations for many problem-solving tasks, and not just because they reduce memory load. Tversky and Suwa showed that architects who sketched during early design phases generated more creative solutions than those who worked through verbal description or verbal-mathematical analysis. The sketch externalized spatial reasoning in a form that could be inspected, critiqued, and modified without rebuilding the entire mental representation.
Mind mapping leverages this advantage: the spatial layout of nodes and connections encodes relationships that prose would have to state explicitly with relative clauses and connectives. Reading a mind map and reading a paragraph describing the same content engage different processing, and the spatial version tends to support better retention of relational structure.
The key insight from this research is that external spatial representations are cognitive tools, not documentation. They do part of the thinking, not just record thinking that's already been done.
How to Strengthen Visual Thinking
Sketch problems before solving them. The act of drawing a situation — even crudely, with no artistic intent — forces spatial decisions about what relates to what. This constraint often reveals missing information or hidden assumptions that verbal description can paper over. The sketch doesn't need to be accurate; it needs to be explicit.
Practice spatial reasoning tasks deliberately. Solving geometric puzzles, working with three-dimensional models, or navigating from memory trains the mental manipulation of spatial representations. These build the same cognitive machinery you use for visual problem-solving in professional domains.
Build visual vocabularies for your domain. Every field has standard diagrams, schematics, and representations. Learning to read and produce these isn't a stylistic preference — it's access to cognitive tools that professionals in the field use to think. An engineer who can't read a circuit diagram is missing a major cognitive affordance that exists precisely because the visual form does something verbal description cannot.
Use mind mapping as a thinking tool, not a documentation tool. Mind maps produce the most value when constructed as part of active reasoning, not after reasoning is complete. Starting with a blank page and spatially arranging concepts as you think forces visual and spatial cognition into the process.
Alternate representations deliberately. Take a problem you've been working on in words and draw it. Take a diagram and describe it in full sentences. The translation between modes forces you to notice what's expressed in one representation that isn't captured in the other — which is where conceptual gaps and unexplored possibilities typically live.
The analogical reasoning exercise trains a closely related capacity: identifying structural correspondences between different domains. Strong visual thinkers often excel at analogical reasoning because spatial representations make structural relationships easier to inspect, compare, and manipulate than verbal descriptions of the same relationships.
Visual Thinking in the Creative Process
The creative process involves multiple cognitive modes at different stages. Early idea generation benefits from visual and spatial thinking because those modes support broad search with low commitment — a sketch is faster to produce and revise than a prototype, and much cheaper than a fully articulated verbal argument. Later stages of evaluation and refinement typically rely more on convergent, verbal, and analytical reasoning.
Integrating visual thinking deliberately means matching cognitive tools to process stages. Use spatial and visual representations during divergent thinking — when you want to generate combinations without prematurely filtering. Switch to verbal and analytical modes when you need to evaluate and refine specific candidates. The common failure mode is applying convergent, verbal evaluation too early, which kills generative search before it has accessed the spatial and imagistic processing where many creative combinations originate.
If you notice your creative work consistently stays verbal — you plan in words, solve in words, and only translate to visual later — that's worth examining. The translation often doesn't happen at all, leaving the spatial mode untapped for the stages where it provides the most leverage.
Ready to train your creativity? Try science-backed exercises that measure and improve your creative thinking. Start a Free Exercise