What do Einstein’s theory of special relativity, penicillin, and microwaves have in common? They were all born from sudden flashes of understanding, known formally in psychology as insight learning.^1,2 Often called “lightbulb,” “Aha!” or “Eureka!” moments, insight learning refers to the spontaneous solution to a tricky problem, appearing “magically” in our minds like a light turning on.^1,3,4 One moment we’re stumped, the next moment we’re not.

Insight learning is closely associated with Gestalt psychology; both are characterized by their suddenness. We seem to come to solutions effortlessly and unexpectedly, without using other problem-solving approaches such as trial and error, observation, or logical deduction.⁴ Non-insight learning, also referred to as analytic learning, involves gradual, step-by-step strategies while insight learning reveals complete solutions with little to no indication of how we came up with them (creating something of a “duh” moment).^3,4 Think of the classic murder mystery story: a brooding detective is painstakingly reviewing the details of a complex case, scanning the files for the hundredth time before everything just “clicks” and the culprit’s identity suddenly becomes clearer than ever. That’s insight learning (dramatically) at work. 

If you’ve ever experienced insight learning, you’re probably familiar with its emotional effects—how each realization feels. Insight generates positive emotions, such as satisfaction, excitement, or happiness once we find the solution.⁴ We also tend to feel more confident about insight-driven solutions. The emotional dimensions of insight, along with its snappiness, differentiate it from analytic learning, which is often accompanied by feelings of hesitancy and uncertainty.⁴ 

Wouldn’t it be great if answers could simply pop into our heads without struggle? Unfortunately, that’s not always the case. Somewhat frustratingly, insight learning usually occurs when our other problem-solving strategies fail, and we get stuck. This period of struggle before we arrive at the conclusion is often referred to as an impasse.³ Like hitting a roadblock in the street, a figurative obstruction forces us to find alternative “routes,” leading to a mental restructuring of the problem.³ We rearrange pieces of the puzzle in our minds, allowing us to approach the problem with a new perspective. 

While there are multiple proposed models of insight learning, the most common one involves four stages: preparation, incubation, illumination (or insight), and verification.⁵ First, we have to understand the problem—gather all the puzzle pieces and know the picture they’re supposed to form. After we get stuck, the mind often enters a period of incubation in which we subconsciously restructure information. This leads to the moment of insight, after which we verify the solution by putting the pieces together to see if they fit.

Analytic Learning: Using structured, step-by-step approaches to solve problems. Also known as non-insight, analytic learning relies on our past experiences, existing knowledge, and logic to identify solutions.^3,4 

Impasse: A period of stagnation during problem-solving in which our attempts to find a solution are unsuccessful. Insights appear after impasses force us to approach the problem in a novel way.³

Restructuring: Mentally reorganizing information about a problem, shifting the way we understand its components. Restructuring often happens subconsciously, leading to the apparent suddenness of insights.^3,4 

Gestalt Psychology: A school of psychological thought that emphasizes how we perceive objects, stimuli, and problems as unified wholes and relationships rather than independent components.

Default Mode Network (DMN): A collection of interconnected brain areas that are active when the brain is resting and not focused on specific tasks. Linked to mind-wandering, creativity, and consciousness, the DMN has been implicated in restructuring and insight learning.^20,23

Humans have been learning via insight long before we had a name or explanation for it. The phrase “Eureka!” is commonly attributed to Archimedes, an ancient Greek scholar and inventor.⁷ As the story goes, Archimedes was stepping into a bath when he noticed that submerging his body caused the water to rise by an equal volume. Suddenly realizing he could use the same method to measure the volume of any irregularly shaped object, Archimedes reportedly yelled the famous phrase. While now widely understood as a legend, the story of Archimedes demonstrates key properties of what we now call insight—namely, how unexpected and elating an insight can be. 

Although humans have long used insight to make scientific advancements and creative innovations, the formal psychological study of learning and insight can be traced back to the late 19th and early 20th centuries. Insight learning emerged as a response to behaviorism, a school of thought that modeled all behavior on learned experience and training.³ Behaviorists held that all learning could be explained by gradually creating associations with consequences, punishments, and rewards, leaving little room for insight learning.

In 1898, psychologist Edward Thorndike laid the foundation for behaviorism, publishing the results of his now-classic puzzle box experiments.^3,7 These experiments showed that cats learned to press a lever to escape a cage by first displaying random trial-and-error behaviors, eventually succeeding by chance, and then gradually learning to press the lever intentionally. Continuing into the early 20th century, Thorndike’s experiments were an important precursor to the work of psychologists John B. Watson and B.F. Skinner, who later formalized the field of behaviorism.

Around the same time, a new school of thought was emerging: Gestalt psychology. Originated by psychologist Max Wertheimer with significant contributions from fellow psychologists Kurt Koffka and Wolfgang Köhler, Gestalt psychology emphasized understanding pieces of information as wholes that are greater than the sum of their parts. This extended to problems as well. In contrast to behaviorism, Gestalt psychology proposed that we learn by processing problems as a whole and (re)considering the relationships between components.²² Wolfgang Köhler critiqued Thorndike’s work, believing that the experiments failed to replicate real-life scenarios where multiple possible solutions exist for most problems.³ 

In 1925, Wolfgang Köhler was the first to formalize and propose insight as a learning mechanism.^3,9,10 

To illustrate the limitations of Thorndike’s experiments, Köhler devised a number of problems where apes could experiment with a variety of objects, such as boxes and sticks, to access food. Köhler observed that some chimpanzees would struggle to access a banana, momentarily give up, and then suddenly appear to have an idea and successfully use a new approach to solve the problems. While Köhler couldn’t read the chimpanzees’ minds (nor can modern scientists, for that matter), it was clear that the chimpanzees were finding solutions abruptly, evidently using a different approach than the one behaviorists had described. 

Köhler and other Gestalt psychologists theorized that humans have two problem-solving modes. Our default mode is analytic problem-solving, in which we use solutions to past problems to construct new ones.¹⁰ Insight learning is our secondary mode, which requires putting aside our past experiences and pre-conceptions to see the problem in a new way.

Since this dual-process theory, psychologists have continued studying the mechanisms of insight learning in humans and in animals. We know it happens, but how? More recent research suggests that the lines between insight and analytic learning might not be so clear—we likely use a mix in our everyday lives.¹⁰ In 1954, French mathematician Jacques Hadamard proposed the four-stage model—preparation, incubation, illumination, and verification—based on mathematical learning.⁵ Other two- and three-stage models have also been introduced, but this remains the most popular.

Today, scientists continue to study how insight works in the brain, and whether insight is its own process or a subprocess of a more general problem-solving system. Technological advancements in neuroimaging methods are allowing scientists to study the neural correlates of insight, potentially creating a pathway for future technology that can facilitate it.¹¹ Researchers are also investigating how we can harness the elusive powers of insight in certain environments, including developing classrooms and workplaces that support creative and flexible problem-solving.¹² 

Edward Thorndike

An American psychologist whose work on learning in animals laid the groundwork for the field of educational psychology in the early 20th century.¹³ Thorndike’s famous puzzle-box experiments provided evidence for reinforcement learning and the law of effect, a principle created by Thorndike to describe how positive and negative consequences shape our behavior.⁸ 

Wolfgang Köhler

Considered a pioneer of Gestalt psychology, Köhler was a German psychologist who originated insight learning theory in 1925. He also notably studied psychophysical isomorphism, the idea that the structure of our brain activity directly correlates with the structure of our subjective, perceptual experiences.

Jacques Hadamard

A renowned French mathematician who made significant contributions to number theory, complex analysis, and other mathematical disciplines. Hadamard also studied the psychological and philosophical dimensions of mathematics, emphasizing the role of insight learning in mathematical breakthroughs and originating the popular four-stage model of insight learning.¹⁴

We’ve likely all benefited from moments of insight, whether it be our own or someone else’s. Insight is a key process that facilitates everything from innovative interdisciplinary research to humor to some of humanity’s greatest inventions. 

Transdisciplinary learning and research

Despite becoming quite the buzzword in recent years, “interdisciplinary” or “transdisciplinary” collaboration has never been more necessary to tackle some of humanity’s greatest challenges, such as climate change. Although interdisciplinary is often an easy sell in theory, putting it into practice can be much more difficult. Various research disciplines can use vastly different conventions, language, and even ways of thinking. After years—perhaps even decades—of training and experience, these can be hard to unravel. How can we overcome those divisions to work with other disciplines, rather than simply alongside them?

In a 2022 article, a group of Swiss researchers in various fields—including technology, environmental sciences, and waste management— theorized that insight is the key to bridging interdisciplinary gaps.¹² They conducted a study on a group of master's students, PhD students, and postdoctoral researchers who were living and working together to develop a community interaction event for residents of villages in rural Switzerland. 

By investigating the participants’ collaborative process, the researchers discovered that “Aha!” moments allowed participants to restructure their thinking while incorporating new information from different disciplines, making the novel connections that are so valuable in interdisciplinary research. In their paper, they posit that the restructuring involved in insight is a key process in integrating knowledge and allowing diverse perspectives to influence the ways we see the world. This process is “transformative,” changing not only the research but the researcher as well.

Realizing the limitations of our mental models can also be uncomfortable. Creating environments where students, researchers, or any individuals feel safe to navigate this restructuring is essential for effective collaboration. This research shows us that, despite being a very subjective experience, insight is heavily influenced by social context. Hence, understanding insight as a social process may help us promote it in our own collaborative spaces.

Cutting-edge inventions

In popular culture, “Eureka!” and “lightbulb” moments are most often associated with genius inventors, brilliant scientists, and talented creatives. And with good reason, too—insight learning has fueled many of history’s greatest innovations. Penicillin, the revolutionary antibiotic that won its inventors the 1945 Nobel Prize in Medicine, was discovered by accident.² One day, Sir Alexander Fleming noticed a clear spot on his petri dish of bacteria where a stray spore of Penicillium mold had landed, spurring a moment of insight in which he realized penicillin’s potential as an antibiotic. This single moment of ingenuity would go on to save millions of lives and transform the treatment of bacterial infections forever.^2,15

Insight has not only changed the landscape of hospitals and clinics around the world, but also your home. In the 1940s, American engineer Percy Spencer was working on a military radar device called a magnetron when he noticed a chocolate bar had melted in his pocket from the heat emitted by the device.¹⁵ This prompted a mental shift for Spencer, making him wonder if the technology could be used to cook food. A couple of heated eggs and popped corn kernels later, Spencer had just invented the microwave, changing restaurants, break rooms, and dorms around the world.

Getting the joke

Have you ever been told a joke that took a moment to settle in? Perhaps a stupid pun or a clever word play? That moment of realization, when things click and you finally get the joke, is a common example of insight at work.^16,17 In some ways, jokes are like problems we can solve; finding the connection between the setup and the punchline often takes time, thought, and a new perspective on language. Rather than working through them algorithmically, we likely experience a sudden illumination where everything makes sense, and we finally understand the humor. 

Consider this joke, for example: “What did the fish say when he ran into a wall?”

“Dam.”

Whether or not you find this particular joke funny is extremely subjective (I won’t be offended if you didn’t), but understanding jokes like this involves high-level cognitive processes like language comprehension and resolving ambiguity.^16,17 Insight allows us to restructure our understanding of the joke—in this case, realizing that the exclamation “damn” could also refer to the underwater wall “dam.” Insight can also explain why humor can bring us so much joy and pleasure. Funny jokes (good or bad) aren’t just clever; they make us feel good.

Although insight can lead to brilliance, it’s not foolproof. Insights are unpredictable and can create excessive confidence and self-assuredness, potentially leading to erroneous conclusions and poor judgment. Further, researchers have conflicting theories about how insight learning works in relation to our broader problem-solving processes.

Misplaced confidence

While realizations reached via insight tend to be correct, they aren’t infallible. Moreover, the characteristic feelings of certainty that accompany insight might lead us astray—we might overly interpret those gut feelings as evidence of truth, leading to false beliefs and weaker critical thinking. In a 2020 study, researchers from Vrije Universiteit Amsterdam, University of California, Santa Barbara, and the University of Queensland tested the idea that “Aha!” moments can mislead us.¹⁸ 

The researchers came up with an experiment to artificially induce insight: participants were presented with a statement and asked to rate its truthfulness (for example, “there are more than 100,000 craters on the moon”). Some participants received the whole passage, while others received only part of the sentence with a keyword scrambled. The researchers hypothesized that participants would experience an “Aha!” moment when they successfully unscrambled the anagram, mistakenly attributing the insight to the truth of the statement. For instance, a participant would realize the anagram “nomo” stood for “moon,” and would think that insight was them confirming there are more than 100,000 craters on the moon. 

The study found that participants who solved the anagram rated the statements as more truthful, even when they were false. Additionally, participants who reported experiencing an “Aha!” moment when they solved the anagram rated statements highest in truthfulness. These results indicate that feelings of insight influence our judgment, potentially serving as a shortcut for logic and reasoning. We might easily fall victim to this phenomenon if we’re not aware of it. As the authors speculate, advertising and media may leverage similar riddles or tricks to mislead audiences or persuade consumers. 

Unpredictable and unreliable

One of the main limitations of insight learning is its unpredictability and unreliability. Since insight inherently obscures the reasoning process and often happens outside of our conscious awareness, it’s difficult to control when we experience insight and why. After all, how can we expect to find an unexpected perspective? 

Recent research may be bringing us closer. Cognitive neuroscientists are using neuroimaging technologies such as electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) to trace insight to specific brain areas and patterns of neuronal activity, with the implication that future technology could stimulate similar activity to facilitate insight.¹¹ Simultaneously, educational researchers are investigating ways we can build learning environments and workspaces that promote insight, facilitating the flexibility and rest necessary to subconsciously restructure our thinking.¹² For example, fostering psychological safety and formalizing breaks might boost restructuring and open sharing of different perspectives. While these efforts continue to advance our understanding of insight, we still have yet to fully predict and control when insight happens. 

Mysterious mechanisms

Today, researchers are still studying how insight works in the brain, with contrasting theories about its underlying mechanisms. The question remains: is insight a separate cognitive process, or is it part of a broader problem-solving system? Two of the most prominent theories are the representational change theory (RCT; later transformed into what is now called redistribution theory) and the progress monitoring theory (PMT; later transformed into the criterion for satisfactory progress theory).⁴ RCT posits that impasses correct our assumptions about what knowledge is necessary for a problem, like past experiences or facts, leading us to restructure our thinking and subconsciously draw from different knowledge. Conversely, PMT proposes that we consciously monitor our problem-solving progress, looking for alternative solutions when certain progress criteria aren’t being met. 

A 2003 study testing these theories yielded evidence to support both, yet leaning in favor of RCT.¹⁹ However, there’s still no clear winner. In 2014, cognitive psychologist Robert Weisberg put forth a new four-stage model of insight that integrates components from RCT and PMT:¹⁰ 

• Stage 1: We first search our long-term memory for similar problems and solutions that could be applied to the new problem (Non-insight).
• Stage 2: If the problem isn’t solved in stage 1, we apply more complex mental shortcuts, or heuristics, to try to find a solution (Non-insight). 
• Stage 3: New information might arise if we fail to solve the problem in stages 1 and 2. We can use this new information to reconceptualize the problem and solve it through insight. If no new information is available, we reach an impasse.
• Stage 4: If we still can’t solve the problem after stages 1-3, we can redistribute the knowledge we’re drawing from to reach an insight.

Weisberg’s integrated model combines the step-wise monitoring system of PMT (essentially evaluating whether or not a stage worked and progressing to the next if it didn’t) with the redistribution of what knowledge is being “activated” as described by RCT. Even still, the integrated theory isn’t conclusive. A 2021 Belgian study could only partially support Weisberg’s integrated theory, suggesting the lines between non-insight and insight problem-solving aren’t as clear as Weisberg thought.⁴ Unfortunately, understanding insight is not nearly as “effortless” as understanding by insight. 

Shower thoughts

Have you ever found the answer to a problem that has plagued you for hours, days, or weeks while in the shower? Or thought of the world’s next greatest invention that would make you millions? Or even discovered the meaning of life in a moment of enlightenment? You’re not alone. Anecdotes of having insights while bathing have become so popular that “shower thoughts” is now a common phrase in many people’s vocabularies. 

The science behind shower thoughts actually reveals interesting details about how our brains work and solve problems. The key is relaxation. When we’re not actively focusing on a task, we activate a group of brain areas whose joint activity forms what we call the default mode network (DMN).²⁰ This network is associated with a resting state, turning on when we do more passive tasks, mind-wander, or relax. A significant body of research has linked the DMN to creative thinking and generating ideas, allowing us to explore information freely before other brain networks can evaluate a solution’s feasibility through logic and rationality.²³ 

Problem-solving and decision-making require a careful balance between creativity and reason—the DMN works in tandem with other brain networks to create solutions that are implementable and effective. However, taking time to activate the DMN when you’re stuck can help facilitate insight by creating space for creative exploration and restructuring. John Kounios, a cognitive neuroscientist and co-author of The Eureka Factor: Aha Moments, Creative Insight, and the Brain, says showers are a great place to do just that.²⁰ Showers are relaxing, and they often don’t take up much mental effort, allowing your brain to make new connections between existing information. So, the next time you can’t find the solution to your problems, you might just find the answer in your shower. 

Expectations vs. reality: realistic program design

Things don’t always go according to plan, no matter how much experience you have. When a group of faculty members sat down to develop a new executive master’s of business administration (EMBA) program at Boise State University, they realized the process of researching and designing a new program wasn’t as tidy as they expected.²¹ The team members documented and later reflected on the unexpected “Aha!” moments that shaped their process, opening a window into how insights influence collaborative design. 

Among the problems they faced, the authors describe one of branding: after surveying competing EMBA programs in the area, the team realized that they were competing with prominent, “BIG-name” programs that were attracting prospective students in the surrounding areas. Their university, which comparatively had weaker branding and was less renowned, stood little chance against the likes of UCLA or Wharton. Amidst their despair (their words), this impasse of competitiveness led to the insight that without a visible brand, their program had an opportunity to define a new image and offer something different to its students. 

By restructuring their branding (or lack thereof) from a weakness to a strength, the team turned toward experiential learning and building connections with the local community, setting them apart from competitors and creatively embracing obstacles. Rather than using their anticipated step-by-step approach to mirror existing programs, the EMBA development team embraced perspective shifts to find novel, creative solutions to big obstacles.

Observational Learning

Since the rise of behaviorism in the early 20th century, psychological research has uncovered more learning mechanisms outside of the behaviorist reinforcement learning model. This article dives deeper into observational learning, a social learning process that involves modeling the behavior of others. 

Creativity in the Workplace: How to bolster engagement and productivity at work 

While researchers are still looking for ways to definitively stimulate insight, there are plenty of proven strategies for facilitating creative thinking. In this article, Janessa Pong discusses how we can build creative workplaces and unlock productivity. 

1. Kounios, J., & Beeman, M. (2015). The eureka factor: Aha moments, creative insight, and the brain. Random House.
2. Tan, S. Y., & Tatsumura, Y. (2015). Alexander Fleming (1881-1955): Discoverer of penicillin. Singapore medical journal, 56(7), 366–367. https://doi.org/10.11622/smedj.2015105 
3. Shettleworth, S. J. (2012). Do animals have insight, and what is insight anyway? Canadian Journal of Experimental Psychology = Revue Canadienne de Psychologie Experimentale, 66(4), 217–226. https://doi.org/10.1037/a0030674
4. Stuyck, H., Aben, B., Cleeremans, A., & Van den Bussche, E. (2021). The Aha! moment: Is insight a different form of problem solving? Consciousness and Cognition, 90, 103055. https://doi.org/10.1016/j.concog.2020.103055
5. Hadamard, J. (1954). An essay on the psychology of invention in the mathematical field. Dover.
6. Sagan, C. (1979). Broca’s brain : reflections on the romance of science (First edition). Random House.
7. Schwarcz, J. (2022, March 5). Is it true that Archimedes formulated his famous principle based on an observation he made as he immersed himself in a bath?. McGill Office for Science and Society. https://www.mcgill.ca/oss/article/history/it-true-archimedes-formulated-his-famous-principle-based-observation-he-made-he-immersed-himself 
8. Chance P. (1999). Thorndike's puzzle boxes and the origins of the experimental analysis of behavior. Journal of the experimental analysis of behavior, 72(3), 433–440. https://doi.org/10.1901/jeab.1999.72-433
9. Kohler, W. (1925). The mentality of apes. (E. Winter, Trans.). Harcourt, Brace.
10. Weisberg, R. W. (2014). Toward an integrated theory of insight in problem solving. Thinking & Reasoning, 21(1), 5–39. https://doi.org/10.1080/13546783.2014.886625
11. Kounios, J., & Beeman, M. (2009). The Aha! moment: The cognitive neuroscience of insight. Current Directions in Psychological Science, 18(4), 210–216. https://doi.org/10.1111/j.1467-8721.2009.01638.x
12. Pearce, B. J., Deutsch, L., Fry, P., Marafatto, F. F., & Lieu, J. (2022). Going beyond the AHA! moment: insight discovery for transdisciplinary research and learning. Humanities and Social Sciences Communications, 9(1). https://doi.org/10.1057/s41599-022-01129-0
13. The Editors of Encyclopaedia Britannica (2024, August 27). Edward L. Thorndike. Encyclopedia Britannica. https://www.britannica.com/biography/Edward-L-Thorndike
14. Cartwright, M. L. (1965). Jacques Hadamard, 1865-1963. Biographical Memoirs of Fellows of the Royal Society, 11, 75–99. https://doi.org/10.1098/rsbm.1965.0005 
15. Ackerman, E. (2024, January 11). A brief history of the microwave oven. IEEE Spectrum. https://spectrum.ieee.org/a-brief-history-of-the-microwave-oven 
16. Canestrari, C., Branchini, E., Bianchi, I., Savardi, U., & Burro, R. (2018). Pleasures of the mind: What makes jokes and insight problems enjoyable. Frontiers in Psychology, 8, 2297. https://doi.org/10.3389/fpsyg.2017.02297
17. Tian, F., Hou, Y., Zhu, W., Dietrich, A., Zhang, Q., Yang, W., Chen, Q., Sun, J., Jiang, Q., & Cao, G. (2017). Getting the joke: Insight during humor comprehension - evidence from an fMRI study. Frontiers in Psychology, 8, 1835. https://doi.org/10.3389/fpsyg.2017.01835
18. Laukkonen, R. E., Kaveladze, B. T., Tangen, J. M., & Schooler, J. W. (2020). The dark side of Eureka: Artificially induced Aha moments make facts feel true. Cognition, 196. https://doi.org/10.1016/j.cognition.2019.104122
19. Jones G. (2003). Testing two cognitive theories of insight. Journal of Experimental Psychology. Learning, memory, and cognition, 29(5), 1017–1027. https://doi.org/10.1037/0278-7393.29.5.1017
20. Colino, S. (2022, August 12). The science of why you have great ideas in the shower. National Geographic. https://www.nationalgeographic.com/magazine/article/the-science-of-why-you-have-great-ideas-in-the-shower 
21. Napier, N. K., Bahnson, P. R., Glen, R., Maille, C. J., Smith, K., & White, H. (2009). When “Aha Moments” make all the difference. Journal of Management Inquiry, 18(1), 64–76. https://doi.org/10.1177/1056492608319679
22. Vitello, M., & Salvi, C. (2023). Gestalt's Perspective on Insight: A Recap Based on Recent Behavioral and Neuroscientific Evidence. Journal of Intelligence, 11(12), 224. https://doi.org/10.3390/jintelligence11120224
23. Luchini, S. A., Volle, E., & Beaty, R. E. (2025). The role of the default mode network in creativity. Current Opinion in Behavioral Sciences, 65. https://doi.org/10.1016/j.cobeha.2025.101551