In the winter of 1865, the German chemist August Kekule was dozing in a chair when he dreamed of a snake biting its own tail — an ancient symbol called the ouroboros. He woke with a start and recognised, immediately, that the image had given him the answer to a problem he had been working on for months: the molecular structure of benzene. The carbon atoms, he realised, must be arranged in a closed ring rather than a linear chain. It was one of the most consequential structural discoveries in the history of organic chemistry, and it came not from deliberate experimental reasoning but from a half-sleep state in which conscious analytical thought had been suspended.
Kekule's experience is one of the most famous examples of creative insight arriving outside conscious effort, but it is not unique. Archimedes and the bathtub. Newton and the apple. Poincare and the mathematical insight that arrived the moment he stepped onto a bus, having thought about nothing. The pattern is too consistent across history and across domains to be coincidence. Creative breakthroughs reliably appear when deliberate conscious effort is released, often after sustained prior engagement with the problem. For most of the history of thinking about creativity, this was noted but unexplained — a mystery attributed to inspiration, genius, or the Muse.
Neuroscience in the past two decades has begun to explain it. The machinery Kekule was running without knowing it — the spontaneous associative processes that fire when analytical vigilance relaxes — has been mapped. It operates through a large-scale brain network called the default mode network. Understanding how that network works, how it interacts with other systems, and what conditions promote or suppress its creative function, provides a foundation for practical creativity that goes substantially beyond the usual advice about thinking outside the box.
"The creative brain is not a special brain. It is a well-connected brain — one in which regions that normally do not communicate have learned to talk to each other." -- Rex Jung, neuroscientist, University of New Mexico, 2014
Key Definitions
Default mode network (DMN): A set of interconnected brain regions — including the medial prefrontal cortex, posterior cingulate cortex, and angular gyrus — that show increased activity during wakeful rest, mind-wandering, and internally-directed thought. Initially described by Marcus Raichle in 2001 as the "default" state of the brain when not engaged in external tasks, the DMN is now understood to be centrally involved in imagination, autobiographical memory retrieval, mental simulation of future events, and the spontaneous generation of ideas.
Executive control network (ECN): A network centred on the dorsolateral prefrontal cortex and posterior parietal cortex that supports goal-directed attention, working memory, and cognitive control. The ECN is associated with deliberate, analytical thinking and was previously assumed to be anticorrelated with the DMN — typically, when one is active, the other is suppressed.
| Brain Network | Role in Creativity | Active When |
|---|---|---|
| Default Mode Network (DMN) | Spontaneous idea generation, daydreaming, associative thinking | Mind wandering, incubation, imagining possibilities |
| Executive Control Network (ECN) | Evaluating and refining ideas, focused attention | Critical assessment, deliberate elaboration |
| Salience Network | Switching between DMN and ECN | Transitioning between generative and evaluative modes |
Salience network (SN): A network centred on the anterior insula and anterior cingulate cortex that monitors internal and external environments for significant information and regulates transitions between the DMN and ECN. In Beaty's three-network model of creativity, the salience network mediates between spontaneous idea generation (DMN) and deliberate evaluation and elaboration (ECN).
Divergent thinking: The ability to generate multiple possible responses to an open-ended problem, associated with creativity research since Joy Paul Guilford's 1950 address to the American Psychological Association. Divergent thinking is assessed through tasks like the Alternative Uses Test (generating many uses for a common object) and contrasts with convergent thinking, which seeks a single correct answer.
Associative hierarchy: A concept from Sarnoff Mednick's 1962 theory of creative thinking, which proposed that creative individuals have flatter associative hierarchies — their mental associations spread more evenly across many possible connections — while less creative individuals have steeper hierarchies, with a few dominant associations that crowd out more distant ones.
The Default Mode Network: The Brain's Creativity Machine
When Marcus Raichle at Washington University published "A Default Mode of Brain Function" in the Proceedings of the National Academy of Sciences in 2001, the finding seemed paradoxical. Brain imaging studies had consistently shown a network of regions that was more active when participants were doing nothing than when they were performing specific tasks. Rather than a resting state, the brain appeared to maintain an organised, active default mode of operation.
Subsequent research revealed what this default mode was doing. Jonathan Smallwood and colleagues showed that the DMN generates the content of mind-wandering — the spontaneous thoughts, images, memories, and mental simulations that arise without deliberate intention. Randy Buckner at Harvard, in a 2008 review in the Annals of the New York Academy of Sciences, synthesised evidence that the DMN functions as a simulation system: it generates imagined future scenarios, retrieves autobiographical memories, and models the mental states of other people. These are all, fundamentally, forms of internally-directed creative thought.
The connection between DMN activity and creativity was hypothesised for years before it was measured directly. The confirmation came through a series of neuroimaging studies using functional connectivity analysis — measuring which brain regions show correlated activity patterns, indicating functional communication between them.
Rex Jung, a neuropsychologist at the University of New Mexico, conducted some of the earliest work documenting structural brain differences associated with creative ability, published beginning around 2010. Jung and colleagues found that higher creative ability was associated with lower cortical thickness in regions associated with focused attention and cognitive control, and higher structural connectivity in regions associated with associative thinking and imagination. The findings suggested that highly creative individuals were, in some sense, better at suppressing the focused-attention circuits that constrain associative range — allowing the DMN to run more freely.
Roger Beaty's Three-Network Model
The most comprehensive neurological model of creative cognition comes from Roger Beaty at Harvard University and Penn State, whose research identified the simultaneous co-activation of the DMN, ECN, and salience network as the neural signature of creative thinking.
Beaty's 2016 paper in NeuroImage and a landmark 2018 study in the Proceedings of the National Academy of Sciences scanned participants performing the Alternative Uses Test and other divergent thinking tasks. The striking finding was that high-creativity responses — those rated as novel and useful by independent judges — were associated with simultaneous activation of the DMN and the ECN. The DMN generated associative content; the ECN evaluated and elaborated on that content; the salience network coordinated the dynamic interaction between the two.
This co-activation pattern was unusual because the DMN and ECN are typically anticorrelated in neuroimaging studies — they function in opposition, with focused attention suppressing mind-wandering and mind-wandering disrupting focused attention. In highly creative individuals and during moments of creative thought, this anticorrelation relaxed: the two networks communicated simultaneously rather than alternating.
Beaty found that the strength of functional connectivity between these three networks — measured at rest, before any creative task — was the strongest neural predictor of creative performance, explaining approximately 10 to 17 percent of variance in creative output across different tasks and measures. This finding suggests that the capacity for creative thought is partly a trait — some individuals have brains that are structurally better connected for simultaneous DMN-ECN communication — but the between-network connectivity is also responsive to mental states, training, and environmental conditions.
The practical implication is that creativity is not simply "thinking freely" (pure DMN activity) or "thinking analytically" (pure ECN activity). It is the ability to hold both simultaneously: generating associative content freely while maintaining the working memory and attentional control needed to evaluate and develop promising ideas. The shower works not because thinking stops but because the analytical suppression of the DMN relaxes while sufficient cognitive function remains to recognise promising associations when they arise.
The Incubation Effect
The phenomenon that Kekule experienced — insight arriving after a period of not consciously working on a problem — has been studied systematically since Graham Wallas described the four stages of creative thought (preparation, incubation, illumination, verification) in The Art of Thought in 1926. The incubation stage — the period of apparent non-work between focused engagement and sudden insight — has been the most scientifically puzzling.
Ap Dijksterhuis and Teun Meurs at Radboud University published a series of studies on "unconscious thought theory" in the Journal of Experimental Social Psychology (2006) demonstrating that participants who were distracted after studying a complex problem outperformed both those who deliberated consciously and those who decided immediately. The distraction period allowed processing that conscious deliberation interrupted.
The mechanism remains debated. One hypothesis, consistent with the DMN research, is that incubation allows the default mode network to run associative searches through memory without the constraints imposed by analytical focus — effectively running background processing on the problem while consciousness is directed elsewhere. A second hypothesis, developed by Sarnoff Mednick's remote associates theory, is that spreading activation in memory — associations spreading from the problem to connected concepts, and from those concepts to further connected concepts — reaches distant memory regions during incubation that focused thinking would not access.
For practical creativity, the incubation research suggests that deliberately building rest periods into creative work — rather than grinding continuously through a difficult problem — is a neurologically grounded strategy, not avoidance. The incubation works only if preceded by genuine focused engagement with the problem: you must load the problem into memory before the background processing can operate on it. The sequence is engagement, incubation, return to the problem.
Mednick, Guilford, and the Structure of Creative Thinking
Joy Paul Guilford's 1950 presidential address to the American Psychological Association introduced divergent thinking as the cognitive capacity most associated with creativity, distinct from the convergent thinking measured by IQ tests. Guilford's alternative uses tasks and similar measures became the foundation of creativity assessment for the following decades.
Sarnoff Mednick at the University of Michigan added a cognitive architecture to Guilford's observations in his 1962 theory of creative thinking, published in Psychological Review. Mednick proposed that creativity resulted from the formation of associative elements into new combinations that met a specific requirement or were useful in some way — and that the critical variable was the structure of an individual's associative hierarchy.
Most people have steep associative hierarchies: when given a stimulus word, a small number of dominant associations come to mind rapidly, crowding out more distant associations. Highly creative people have flatter hierarchies: their initial associations are less dominant, allowing attention to spread to more distant and unexpected connections. The Remote Associates Test, developed by Mednick, measures the ability to find a common associate linking three seemingly unrelated words — a task that requires reaching across the full range of associative memory rather than settling for dominant responses.
The relationship between Mednick's associative hierarchy theory and Beaty's neural co-activation model is direct: a flatter associative hierarchy corresponds to a more active and less constrained DMN, while the ECN's role is to evaluate which of the many associations generated are actually useful. Highly creative performance requires both the broad associative generation and the selective evaluation — neither alone is sufficient.
Adam Grant, Procrastination, and the Courage to Create
Social psychologist Adam Grant at the Wharton School published a study in 2016 examining the relationship between procrastination and creative performance in a business setting. Working with colleagues, Grant found that moderate procrastination on creative tasks — starting a project, stepping away before completing it, allowing the problem to incubate — produced more creative outcomes than either immediately completing the task or extreme procrastination that left insufficient time for execution.
The finding challenges the conventional association between procrastination and poor performance. In creative domains, the incubation period that procrastination provides can generate more novel and useful solutions than immediate completion. The mechanism Grant proposed was consistent with the incubation research: the delay period allows the DMN to continue processing the problem without the constraints of active deliberation.
Grant's related research on originals — people who champion novel ideas in organisations — found that creative courage was a more consistent predictor of creative contribution than creative ability. People who had ideas but waited for certainty before acting, or who self-censored based on social evaluation concerns, contributed less than people who were willing to propose half-formed ideas and develop them through iteration. The implication is that the social and psychological conditions that support creative contribution — psychological safety, low evaluative threat, tolerance for early-stage imperfection — are as important as the neurological conditions for creative thought.
Constraints as Creativity Catalysts
One of the most counterintuitive findings in creativity research concerns the effect of constraints on creative output. The intuitive model holds that creativity requires freedom: remove restrictions, provide open-ended latitude, and creative thinking will flourish. The research evidence is substantially more nuanced.
Patricia Stokes at Columbia University studied the development of creativity in the careers of artists including Monet, Picasso, and Cezanne, publishing her analysis in Creativity from Constraints (2006). Her finding was that the most creative periods in each artist's career were those in which they had imposed significant constraints on their work — limiting palette, limiting subject matter, limiting technique. The constraints forced novel solutions that would not have emerged from unconstrained work.
Catrinel Haught-Tromp at Rider University documented similar effects in language: people asked to compose sentences using specific improbable word pairings produced more creative outcomes than those given open-ended prompts. The constraints narrowed the solution space in ways that forced exploration of less obvious regions.
The neurological mechanism may involve the ECN: constraints activate goal-directed attention, which channels the DMN's associative generation in productive rather than diffuse directions. Completely open-ended tasks may produce such broad associative spread that no particular direction receives sufficient developmental attention. Constraints serve as focusing mechanisms that direct the associative search without entirely suppressing it.
For practical application, this suggests that working with constraints — self-imposed deadlines, limitations on materials or tools, specific problem framings — can enhance rather than reduce creative output, particularly when the constraints are specific enough to direct attention but not so tight as to prevent novel solutions.
Environment, Colour, and the Physical Conditions of Creativity
Ravi Mehta and Rui Zhu's 2009 research at the University of British Columbia, published in Science, examined the effect of environmental colour on cognitive performance. Their five studies found a consistent pattern: red environments enhanced performance on detail-oriented tasks requiring accuracy and vigilance, while blue environments enhanced performance on creative tasks requiring novel association. The effect was attributed to associative priming — red activates avoidance associations (danger, stop signals) that focus attention, while blue activates approach associations (sky, ocean, openness) that facilitate expansive thinking.
Juliet Zhu at the University of British Columbia documented similar effects of ceiling height, published in the Journal of Consumer Research in 2007. Rooms with higher ceilings produced better performance on tasks requiring abstract, free-associative thinking, while lower-ceilinged rooms produced better performance on tasks requiring focused, concrete thinking.
Mehta, Zhu, and Cheema's 2012 study in the Journal of Consumer Research found that ambient noise at moderate levels — approximately 70 decibels, the level of a typical coffee shop — enhanced creative performance compared to silence. The effect was mediated by a moderate increase in processing difficulty that induced diffuse, abstract thinking without the severe distraction of high noise levels.
The cumulative environmental research suggests that the physical and sensory context of creative work is not trivial. Open spaces, moderate ambient noise, blue and natural colour environments, and moderate temperature produce measurable improvements in creative performance relative to closed, silent, red, or extreme-temperature environments. These are manageable variables for most people and represent low-cost, high-leverage interventions in creative productivity.
Csikszentmihalyi's Flow State and Creative Performance
Mihaly Csikszentmihalyi's research on flow — the state of intense, effortless engagement in a challenging activity that he documented through interviews with artists, scientists, athletes, and surgeons in the 1970s and 1980s, published systematically in Flow: The Psychology of Optimal Experience (1990) — describes a mental state that is both deeply creative and deeply satisfying.
Flow occurs at the intersection of high challenge and high skill: the task is sufficiently demanding to require full engagement but sufficiently within reach that it feels achievable with concentrated effort. When challenge significantly exceeds skill, anxiety dominates. When skill significantly exceeds challenge, boredom results. Flow lives in the channel between them.
Neurologically, Arne Dietrich at the American University of Beirut proposed in a 2004 paper in Consciousness and Cognition that flow involves a transient hypofrontality — a reduction in prefrontal cortex activity that reduces self-monitoring, critical self-evaluation, and analytical inhibition of action. This is consistent with the DMN research: in flow, the executive monitoring that normally suppresses associative thinking is transiently reduced, allowing fluid, automatic performance that integrates high skill with unrestricted associative generation.
For creative work, the flow research suggests that the conditions supporting it — a specific challenging goal, immediate feedback on performance, a skill level matched to the challenge, and an environment free of interruption — are worth engineering deliberately. The creative productivity of a two-hour flow state typically exceeds that of a full day of fragmented, distracted work at similar nominal effort.
Practical Takeaways
Build deliberate incubation into your creative process. When stuck on a difficult creative problem, stop working on it and do something routine and undemanding: walk, shower, cook, do light exercise. Return to the problem after the incubation period. This is not procrastination — it is neurologically grounded problem-solving.
Protect the conditions for simultaneous DMN-ECN activity. Excessive stress, time pressure, evaluative threat, and exhaustion all suppress the relaxed alertness needed for creative insight. Schedule creative work during your highest-energy periods and protect that time from interruptions.
Sleep before returning to unsolved creative problems. REM sleep specifically reorganises memory in associative ways that produce insights unavailable to waking analytical thought. Problems that remain unsolved at the end of a day are good candidates for overnight incubation.
Use constraints deliberately. When creative work feels too open-ended and direction is lacking, impose a specific constraint: limit yourself to three colours, five hundred words, a single tool, a single material. Observe how the constraint forces solutions that unconstrained work would not produce.
Experiment with environment. For creative work, prefer open spaces with moderate ambient noise, natural light, and blue or green tones when possible. Reserve focused analytical work for quieter, higher-vigilance environments.
Generate before evaluating. Premature evaluation kills associative range. Establish explicit phases of divergent generation (no evaluation allowed) and convergent evaluation (critical analysis of generated ideas). Mixing the two phases within a single session reliably reduces creative output.
Expand associative raw material. Read broadly across unrelated fields. Develop interests that have nothing to do with your primary domain. Maintain a diverse network. The brain makes connections between existing memories; the richness of creative output is partly determined by the richness and diversity of the knowledge structures available to connect.
References
- Beaty, R. E. et al. "Robust Prediction of Individual Creative Ability from Neural Activity." Proceedings of the National Academy of Sciences, Vol. 115, No. 5, 2018.
- Beaty, R. E. et al. "Creativity and the Default Network: A Functional Connectivity Analysis of the Creative Brain at Rest." NeuroImage, Vol. 116, 2016.
- Buckner, R. L. et al. "The Brain's Default Network: Anatomy, Function, and Relevance to Disease." Annals of the New York Academy of Sciences, Vol. 1124, 2008.
- Dijksterhuis, A. & Meurs, T. "Where Creativity Resides: The Generative Power of Unconscious Thought." Consciousness and Cognition, Vol. 15, No. 1, 2006.
- Mednick, S. A. "The Associative Basis of the Creative Process." Psychological Review, Vol. 69, No. 3, 1962.
- Wagner, U. et al. "Sleep Inspires Insight." Nature, Vol. 427, No. 6972, 2004.
- Csikszentmihalyi, M. Flow: The Psychology of Optimal Experience. Harper & Row, 1990.
- Grant, A. M. et al. "The Productive Procrastinator: A Study in Creative Performance." Academy of Management Discoveries, 2016.
- Mehta, R. & Zhu, R. "Blue or Red? Exploring the Effect of Color on Cognitive Task Performances." Science, Vol. 323, No. 5918, 2009.
- Mehta, R., Zhu, R. & Cheema, A. "Is Noise Always Bad? Exploring the Effects of Ambient Noise on Creative Cognition." Journal of Consumer Research, Vol. 39, No. 4, 2012.
- Stokes, P. D. Creativity from Constraints: The Psychology of Breakthrough. Springer, 2006.
- Dietrich, A. "Neurocognitive Mechanisms Underlying the Experience of Flow." Consciousness and Cognition, Vol. 13, No. 4, 2004.
Frequently Asked Questions
What parts of the brain are most active during creative thinking?
Research by Roger Beaty and colleagues, published in the Proceedings of the National Academy of Sciences in 2018, identified a co-activation pattern between three large-scale brain networks during creative thinking that overturned earlier assumptions. The default mode network (DMN) — active during mind-wandering and internally-directed thought, and associated with imagination and autobiographical memory — was found to be simultaneously active with the executive control network (ECN), which handles goal-directed attention and working memory. This co-activation was previously thought to be impossible: the DMN and ECN are typically anticorrelated, meaning that when one is active, the other is suppressed. The salience network (SN) — which monitors the environment and directs attention — mediates between the two, determining what internally generated content (from the DMN) gets passed into working memory for elaboration (by the ECN). Beaty's research found that the strength of functional connectivity between these three networks was the best neural predictor of creative performance scores, explaining approximately 10 to 17 percent of variance in creative output.
Why do great ideas come in the shower?
The shower phenomenon is a real and well-documented experience that reflects the neuroscience of the incubation effect. When you are working on a difficult problem and then stop consciously working on it — in the shower, walking, exercising, or doing a routine non-demanding task — the default mode network, which was suppressed during focused work, becomes active. The DMN generates associative thinking: linking distant concepts, combining information from different memory systems, and following chains of association that goal-directed thought suppresses. Research by Ap Dijksterhuis and Teun Meurs, published in the Journal of Experimental Social Psychology in 2006, demonstrated that participants who were distracted after studying a complex problem made better decisions than participants who deliberated consciously or decided immediately — what the researchers called 'unconscious thought theory.' The shower provides the idle processing state in which the DMN can connect the dots. Additionally, Arne Dietrich at the American University of Beirut has documented that the mild dopaminergic arousal produced by warm water and relaxation can facilitate associative connections without triggering the vigilant analytical state that suppresses them.
Is creativity a talent or a skill?
The evidence from behavioural genetics and training studies suggests creativity is both — and that the distinction matters less than it might appear. Twin studies, including a comprehensive review by Daniel Piffer in 2012, find moderate heritability for creative traits: roughly 20 to 55 percent of variance in creative ability is explained by genetic factors, depending on the measure used. This means that 45 to 80 percent of variance is accounted for by environment, experience, and practice. Research on divergent thinking training — the ability to generate multiple solutions to open-ended problems — consistently finds meaningful improvement through practice. A 2014 meta-analysis by Scott Barry Kaufman and colleagues found significant effects of creativity training on multiple measures of creative performance. The practical implication is that creative capacity is trainable even if there is a genetic component to baseline ability. The specific capacities most amenable to training are divergent thinking, associative fluency, tolerance for ambiguity, and the willingness to generate a large number of ideas before evaluating any of them.
How does sleep enhance creative problem-solving?
Ullrich Wagner and colleagues at the University of Lubeck published research in Nature in 2004 showing that REM sleep specifically enhances creative insight. Their experiment gave participants a mathematical transformation task that had a hidden rule which, once discovered, allowed the problem to be solved much faster. After eight hours of sleep including REM, participants were nearly three times more likely to discover the hidden rule than participants who had been awake for the same eight-hour period. The mechanism is the memory consolidation and reorganisation that occurs during REM sleep: the hippocampus replays experiences and integrates them with existing knowledge structures in the neocortex, and during this integration, non-obvious associations between distant pieces of information are formed. Matthew Walker's research at UC Berkeley further established that REM sleep specifically recombines memories in novel ways, effectively running an associative search through memory space that produces connections that waking analytical thought would not generate. The practical implication is that sleeping on a problem — particularly one that requires insight rather than deliberate reasoning — is not procrastination. It is a neurologically grounded problem-solving strategy.
What environments boost creative thinking?
Ravi Mehta and Rui Zhu's 2009 research at the University of British Columbia, published in Science, found that blue environments enhanced creative performance while red environments enhanced detail-oriented performance. The mechanism proposed was associative priming: blue is associated with openness, sky, and ocean (expansive, exploratory states), while red is associated with danger and stop signals (vigilant, detail-focused states). Separate research by Juliet Zhu and colleagues found that 2.4 metres of ceiling height enhanced creative performance compared to lower ceilings, through a similar mechanism of openness versus confinement priming. Ambient noise at moderate levels (approximately 70 decibels, equivalent to a coffee shop) has been shown by Ravi Mehta, Rui Zhu, and Amar Cheema in a 2012 study in the Journal of Consumer Research to enhance creative performance compared to silence, by inducing a moderate level of distraction that activates more diffuse, associative thinking. Complete silence, the research suggests, may promote excessive focus that constrains creative range.
Does stress kill creativity?
Acute stress reliably impairs creative performance through well-documented neurological mechanisms. Cortisol release during stress activates the prefrontal cortex's threat-response mode, narrowing attention and prioritising fast, analytical, System 2 processing over the broad associative thinking that creativity requires. Teresa Amabile's research at Harvard Business School, published in the Academy of Management Journal, found through diary studies of creative workers that time pressure was one of the strongest negative predictors of creative performance — people felt more creative under deadline pressure but their output showed measurably reduced novelty and quality. However, the relationship between arousal and creativity is not simply negative: research by Mark Davis and colleagues found an inverted-U relationship in which moderate arousal enhanced creative performance relative to both very low and very high arousal states. The ideal state for creative work appears to be relaxed alertness rather than either high stress or drowsiness — a combination produced by conditions including moderate exercise, positive mood, physical comfort, and the absence of time pressure or evaluative threat.
Can you train yourself to be more creative?
Yes, within meaningful limits. The specific creative capacities most responsive to deliberate training are divergent thinking (generating many ideas from a single prompt), associative flexibility (making connections between distant concepts), and cognitive flexibility (switching between perspectives and problem framings). Research by Simone Ritter and colleagues at Radboud University found that merely engaging in unusual activities — performing familiar tasks in unexpected ways, following unexpected behavioural sequences — measurably enhanced subsequent performance on creative thinking tasks, by activating cognitive schemas that suppress the default patterns that constrain creative range. Adam Grant's research on procrastination and creativity, published in 2016, found that moderate procrastination on creative tasks enhanced creative output: people who started a project immediately and completed it quickly produced less creative work than those who started, let the problem incubate, and returned to it after a delay. This suggests that deliberately building incubation time into creative processes is a learnable strategy with documented benefits. More broadly, practices that expand associative range — broad reading across disciplines, exposure to unfamiliar cultures, diverse social networks — increase the raw material available for creative combination.