How We Solve Problems: The Psychology of Thinking

Almost everyone knows the feeling of being stuck on a problem. A maths question that will not yield, a flat-pack wardrobe with a leftover screw, a project that has hit a wall. You try the obvious approaches, they fail, and the harder you push, the more stuck you feel. Then, sometimes, the answer arrives all at once — in the shower, on a walk. The problem has not changed. Something in how you were thinking about it has.

Problem-solving is one of the central topics in cognitive psychology: the study of how the mind perceives, remembers, thinks, and decides. Psychologists have spent more than a century studying how people move from a stuck state to a solved one, and their findings are surprisingly practical. This article explains what research has found about how we solve

This article describes patterns found across many people in many studies, not rules for any one person. It cannot diagnose how you think or predict whether you will solve a particular problem. It can help you understand the mental moves that make problem-solving work — and the habits that quietly get in the way. The studies behind it describe averages across groups of people, not verdicts on individuals.

The Stages of Solving a Problem

Psychologists often describe problem-solving as moving through stages, and the clearest version comes from the mathematician George Pólya, whose book How to Solve It (1945) laid out a sequence researchers still use today. The first stage is understanding the problem: what exactly is being asked, what is known, and what is unknown. This sounds obvious, but it is where many attempts fail — a misunderstood problem is almost impossible to solve, because every later step answers the wrong question. Restating the problem in your own words is one of the most reliable moves there is.

The second stage is devising a plan: breaking the problem into smaller parts, working backward from the goal, looking for a similar problem you have solved before, or drawing a diagram. The third is carrying out the plan, and the fourth is looking back: checking whether the answer works and what it teaches you for next time. Good solvers do not rush these stages. They spend more time than poor solvers on the first two — understanding and planning — and less time on frantic trial and error.

Everyday life is full of these stages, even when we do not notice them. Planning a journey is a roblem: you define the destination and the constraints, consider routes, choose one, and adjust when the train is cancelled. The stages are the same for a crossword, a budget, or a broken appliance — and solvers who pause to define and plan outperform the ones who jump straight to trying things.

Algorithms and Heuristics: Two Kinds of Strategy

When people choose a strategy, they generally use one of two kinds. An algorithm is a step-by-step procedure that guarantees a solution if followed correctly — trying every combination on a lock, or long division. Algorithms are reliable, but they can be slow, sometimes impossibly slow: if a problem has too many possibilities, trying them all would take longer than a lifetime.

A heuristic is a rule of thumb: a shortcut that usually works and is fast, but is not guaranteed. Estimating a crowd by how full the room looks is a heuristic; so is guessing that the restaurant with the longest queue is the best. Heuristics are the mind’s way of coping with problems too big for algorithms, and most everyday problem-solving runs on them.

Heuristics come with a trade-off. They are fast and usually good enough, but they can lead the mind astray — the mental shortcuts behind many of the biases the site’s guide to cognitive biases describes. A solver who always trusts the first plausible answer will be wrong more often than one who pauses to check. Daniel Kahneman’s account of fast and slow thinking — the subject of the site’s review of Thinking, Fast and Slow — makes the same point: quick, intuitive thinking is efficient, but it needs a slower check when the stakes are high. The practical lesson is not to abandon heuristics. It is to know when you are using one, and to slow down for the problems that matter.

Insight: When the Answer Arrives All at Once

Some problems are solved gradually, one step at a time. Others are solved in a flash: you struggle for an hour, give up, and the answer appears whole. Psychologists call this insight, and the classic studies come from Wolfgang Köhler, who watched chimpanzees solve problems in the 1910s and 1920s and described them in The Mentality of Apes (1925).

Köhler’s most famous animal was Sultan, a chimpanzee who needed bananas hanging out of reach. After failing to jump for them, Sultan stacked boxes, climbed up, and took the fruit; on another he joined two short sticks to make a long one and raked in food. Köhler argued that Sultan had not learned these solutions by trial and error. He had suddenly seen the situation differently — the boxes were steps, not seats; the sticks were parts of a tool. Insight is a restructuring: the same elements, seen in a new relationship.

Human insight works in a similar way. Janet Metcalfe and David Wiebe (1987) asked people to solve insight problems (riddles with a sudden “aha” solution) and ordinary problems (like algebra) while rating how close they felt to the answer — their “feeling of warmth” — every few seconds. For ordinary problems, warmth rose steadily. For insight problems, it stayed flat and low until the solution arrived — then jumped. People had almost no sense they were close — and then, suddenly, they were done. That is why insight feels the way it does: the mind reorganises the problem until the answer becomes obvious, instead of gradually approaching it.

Insight is not magic, and it is not random: it tends to arrive after a period of work on the problem. But the feeling of suddenness is real, and it tells us something important — sometimes the barrier is not missing information but a wrong way of seeing what is already there.

Mental Set: When Experience Gets in the Way

If insight is about seeing a problem freshly, its opposite is mental set: the tendency to keep using an approach that worked before, even when it no longer fits. The classic demonstration comes from Abraham Luchins (1942), in experiments with water jars known as the Einstellung experiments — Einstellung being German for “set” or “attitude.”

Luchins gave people problems about measuring out water with jars of different sizes. The first several could all be solved with the same formula: fill the largest jar, then pour off the amounts in the two smaller jars. Participants practised this formula until it became automatic. Then came test problems with a much simpler solution — though the old formula would also work, with extra effort. Most participants kept using the complicated formula, and many missed the simpler solution entirely. Participants never trained on the formula found the simple solution easily.

The lesson is uncomfortable: experience helps, but also narrows. The more practised a method is, the harder it is to set aside, even when a better one is staring you in the face. Mental set shows up everywhere: the programmer who reaches for the same framework for every project, the manager who applies last year’s solution to this year’s problem, the driver who always takes the same route after a faster road opens. None of them are being careless — they are doing what worked, and what worked has become invisible. Breaking a mental set usually requires a deliberate pause: “is there a simpler way?”

Functional Fixedness: Seeing Only One Use

A close cousin of mental set is functional fixedness: seeing an object only in terms of its usual function, and missing other uses. The classic study is Karl Duncker’s candle problem, published in On Problem Solving (1945). Participants were given a candle, a box of tacks, and matches, and asked to fix the candle to the wall so it would burn without dripping wax on the table. The solution is to empty the box, tack it to the wall, and stand the candle in it as a shelf. Most people struggle, because they see the box only as a container for the tacks — not as a platform.

Robert Adamson (1952) replicated the experiment with a small change: in one version the box was presented full of tacks, as a container; in another it was presented empty, beside the tacks. People who saw the empty box solved the problem far more easily. The box was the same object in both cases — what differed was how the solver saw it. Functional fixedness is not about intelligence, but about how the mind organises what it sees: once an object has a role, it sticks.

Functional fixedness is easy to spot in everyday life. A screwdriver is for screws — until you need to open a paint tin and it becomes a lever. A shoe is for walking — until you need a hammer. A paperclip is for paper — until you need to press a reset button. People who solve these quickly are not more creative in general. They are, in that moment, less fixed on the usual function. One reliable way to loosen fixedness is to ask: “What else could this be?”

Incubation: Why Stepping Away Helps

There is a familiar experience that sits oddly beside the research on mental set and fixedness: you struggle with a problem, set it aside — and the answer appears. Psychologists call the break incubation, and for a long time its effects were disputed: some studies found a break helped, others found nothing. The question was settled, as far as it can be, by a meta-analysis from Uttara Sio and Thomas Ormerod (2009) combining 117 experiments on incubation and problem-solving.

Their analysis found a positive incubation effect overall: taking a break genuinely improved the chances of solving a problem. But the effect was not the same for every kind of problem or break. Problems calling for divergent thinking — generating many possible ideas — benefited more than linguistic or visual insight puzzles. Longer preparation before the break helped, and what you do during the break matters: a demanding task reduced the benefit, while a low-demand activity — a walk, a shower, a routine chore — preserved it.

Why does stepping away work? Two explanations stand out. The first is that a break lets the wrong approach fade: the mental set blocking you weakens while you are not rehearsing it, so the problem looks fresher on return. The second is that a break lets new information in: a walk, a chance remark, a different view can supply the missing piece. Incubation is not the unconscious mind solving the problem in secret. It is the ordinary mind, given a chance to drop a bad approach and notice something new. When you are genuinely stuck, the most productive thing you can do is often to stop — provided you have put real work in first, and the break is a real break, not just a different kind of effort.

What Helps: Evidence-Based Strategies

The research in this article points to strategies that are simple to state and genuinely supported by the evidence. None is a trick, and none guarantees a solution. They are habits that make good solving likelier.

Define the problem in your own words. Before trying anything, say what you are trying to do, and what you know and do not know. A problem clearly defined is already half understood — and if you cannot state it plainly, you are probably solving the wrong one.

Break it into smaller parts. A large problem is often several small problems in disguise. Solving the easiest piece first turns an overwhelming task into manageable steps, and each success tells you something about the rest.

Work backward from the goal. Start where you want to end up and ask what would have to be true just before that. It helps most when the goal is clear but the path is not — a natural antidote to mental set.

Change the representation. If words are not working, draw it; if a diagram is not working, describe it aloud. Many problems are hard only in the form they are stated in — changing the form, from numbers to a picture or a story to a timeline, can make the answer visible.

Question the function. When stuck, ask what each thing can physically do, not what it is for. Duncker’s candle problem is solved by exactly this move — the question “what else could this be?” is worth asking out loud.

Take a real break. If you have worked hard and are going around in circles, step away and do something undemanding. The break works best after genuine preparation, and when it is not filled with another demanding task.

Explain it to someone else. Describing a problem to another person — or even to an empty chair — forces you to organise your thinking, and the gaps show as you speak. It also brings fresh eyes: someone with no mental set about your problem may see the simple solution you missed.

None of these strategies requires special talent. They are ordinary moves, powerful when used deliberately rather than leaving problem-solving to habit. The research does not promise that every problem has an answer — but the way you approach a problem changes the odds, and the odds are worth improving.

Key Takeaways

  • Problem-solving moves through stages — understanding, planning, carrying out, looking back — and good solvers spend the most time on the first two (Pólya, 1945).
  • Algorithms guarantee a solution but can be impossibly slow; heuristics are fast rules of thumb that usually work but can mislead, so they deserve a deliberate check.
  • Insight is a restructuring of the problem rather than gradual progress: people feel no closer until it suddenly arrives (Metcalfe & Wiebe, 1987), as Köhler’s chimpanzee studies first showed.
  • Mental set is the tendency to keep using an approach that worked before, even when a simpler one exists — Luchins (1942) showed practised solvers missed the easy solution untrained solvers found at once.
  • Functional fixedness is seeing an object only in its usual role: Duncker (1945) and Adamson (1952) showed the same box is far easier to use as a shelf when it is not presented as a container.
  • Incubation helps, but only under conditions: a meta-analysis of 117 experiments found a real benefit, strongest after thorough preparation and during a low-demand break (Sio & Ormerod, 2009).

References

  • Adamson, R. E. (1952). Functional fixedness as related to problem solving: A repetition of three experiments. Journal of Experimental Psychology, 44(4), 288–291. https://doi.org/10.1037/h0062487
  • Duncker, K. (1945). On problem-solving. Psychological Monographs, 58(5), i–113. https://doi.org/10.1037/h0093599
  • Köhler, W. (1925). The Mentality of Apes. Routledge & Kegan Paul.
  • Luchins, A. S. (1942). Mechanization in problem solving: The effect of Einstellung. Psychological Monographs, 54(6), i–95. https://doi.org/10.1037/h0093502
  • Metcalfe, J., & Wiebe, D. (1987). Intuition in insight and noninsight problem solving. Memory & Cognition, 15(3), 238–246. https://doi.org/10.3758/BF03197722
  • Pólya, G. (1945). How to Solve It. Princeton University Press. https://doi.org/10.1515/9781400828678
  • Sio, U. N., & Ormerod, T. C. (2009). Does incubation enhance problem solving? A meta-analytic review. Psychological Bulletin, 135(1), 94–120. https://doi.org/10.1037/a0014212

Keep Exploring

What is the difference between an algorithm and a heuristic?

An algorithm is a step-by-step procedure that guarantees a solution if followed correctly, like trying every combination on a lock. A heuristic is a rule of thumb that is fast and usually works, like judging a crowd by how full the room looks, but it is not guaranteed and can lead you astray.

What is insight in problem-solving?

Insight is a sudden restructuring of a problem: the same elements are seen in a new relationship, and the answer becomes obvious all at once. Research shows people feel no closer to the answer until it arrives, which is why insight feels sudden rather than gradual.

What is functional fixedness?

Functional fixedness is the tendency to see an object only in terms of its usual function — a box as a container, for example — and to miss other uses, like using it as a shelf. Duncker’s candle problem is the classic demonstration, and asking “what else could this be?” is the usual remedy.

Does taking a break really help you solve a problem?

A meta-analysis of 117 experiments found a real incubation effect, but it depends on the conditions: it helps most after thorough preparation on the problem, and when the break involves a low-demand activity like a walk rather than another demanding task.

Why do smart people get stuck on simple problems?

Usually because of mental set or functional fixedness: a practised approach or a fixed view of an object keeps working in the mind even when a simpler solution is available. Luchins showed that people trained on one method missed the easy answer that untrained people found at once.

W

Waqar holds a Master’s degree in Applied Psychology and works as an SEO content writer. He writes clear, research-informed content that helps readers understand psychology and related topics.

Leave a Comment