Problem-Solving Strategies

Why This Matters

Problem-solving is at the heart of cognitive psychology and one of the most heavily tested areas on exams. You're not just being asked to recall what "means-end analysis" means; you're being tested on when different strategies work best, why some approaches guarantee solutions while others trade accuracy for speed, and how cognitive barriers like functional fixedness prevent us from finding solutions. Understanding these strategies reveals fundamental principles about how the mind represents problems, searches for solutions, and sometimes gets stuck.

The strategies here demonstrate core concepts like algorithmic vs. heuristic processing, problem representation, cognitive flexibility, and the role of unconscious processing. When you study these, focus on the underlying mechanisms: What makes one approach systematic and another intuitive? Why do "aha" moments happen? Don't just memorize definitions. Know what cognitive principle each strategy illustrates and when you'd use one approach over another.

---

Systematic Search Strategies

These approaches involve methodically working through a problem space. They rely on explicit, step-by-step processing and work best when the problem is well-defined, meaning you know the initial state, the goal state, and the available moves.

Means-End Analysis

This strategy works by repeatedly shrinking the gap between where you are and where you want to be.

1. Compare your current state to the goal state and identify the largest difference.
2. Select an action ("operator") that would reduce that difference.
3. If you can't apply that operator directly, create a sub-goal to remove whatever is blocking it.
4. Apply the operator, then return to step 1 with your new current state.

The Tower of Hanoi is the classic example. You can't move the largest disk directly to the goal peg because smaller disks are on top of it. So you set up sub-goals: move the smaller disks out of the way first. Each sub-goal brings you closer to the final arrangement.

Working Backward

This strategy starts from the goal and reverses toward the starting conditions. It's particularly powerful when the end state is clear but there are too many possible first moves to search forward efficiently.

You ask: "What must be true right before the goal is reached?" Then you ask the same question about that state, and keep going until you arrive at your starting point. Mathematical proofs and planning problems often benefit from this approach because the goal constrains which paths actually work.

Forward Chaining

Forward chaining begins with known information and applies rules to derive new conclusions step by step. It's a data-driven approach: you look at what you have, see what rules apply, generate new information, and repeat until you reach the goal.

This contrasts with backward chaining, which starts from a hypothesis and works backward to check whether the data supports it. Humans use forward chaining whenever they reason "if A, then B; if B, then C..." It's also the logic behind many expert systems in AI.

Algorithm

An algorithm is a procedure that guarantees a correct solution if followed precisely. There's no guesswork involved.

The tradeoff is time. Think of solving an anagram by testing every possible letter combination. You'll eventually find the word, but it could take a very long time. This makes algorithms reliable but often impractical for complex problems, which is exactly why heuristics exist.

---

Heuristics and Mental Shortcuts

Heuristics sacrifice guaranteed accuracy for cognitive efficiency. They exploit patterns and rules of thumb that usually work but can lead us astray.

Heuristics

Heuristics are fast, frugal mental shortcuts that reduce cognitive load when making decisions or solving problems. Two of the most important ones come from Tversky and Kahneman's research:

• The availability heuristic judges probability by how easily examples come to mind. You might overestimate plane crash risk because crashes are vivid and memorable, even though driving is statistically far more dangerous.
• The representativeness heuristic judges category membership by similarity to a prototype. You might assume someone who reads poetry is more likely a professor than a truck driver, ignoring the fact that there are far more truck drivers in the population (this is called neglecting base rates).

These heuristics are often accurate, but they produce systematic biases. Understanding when and why they fail is key exam content.

Trial and Error

Trial and error tests solutions unsystematically until one works. There's no guiding strategy, just persistence. If you're locked out and trying random key combinations, that's trial and error.

It's useful when the problem structure is unknown and you can't predict which approach will succeed. It's inefficient, but it can lead to unexpected solutions precisely because it doesn't follow a plan.

---

Insight and Restructuring

These involve sudden shifts in understanding rather than incremental progress. Insight occurs when you restructure your mental representation of a problem.

Insight Problem Solving

Insight is characterized by "aha" moments. Solutions appear suddenly after a period of apparent stuckness, often accompanied by a feeling of certainty.

What's happening cognitively is a restructuring of the problem representation. You aren't making gradual progress; you're jumping from one way of understanding the problem to a completely different one. This is why solvers often can't articulate how they reached the answer, which distinguishes insight from analytical problem-solving.

Representation and Restructuring

How you frame a problem determines what solutions you can see. A poor representation blocks progress because it makes the correct solution invisible.

The nine-dot problem is a classic example: most people assume the dots form a boundary they can't cross, and only by abandoning that assumption can they find the solution. The same problem can be easy or hard depending entirely on how it's represented. This is one of the most important ideas in problem-solving research.

Restructuring means changing your mental model of the problem, shifting from one problem space to another.

Incubation

Taking a break from a problem can facilitate insight. Stepping away allows unconscious processing to continue working on the problem while you do something else.

Incubation also reduces fixation on unproductive approaches by letting your mental set decay. When you return, you're less locked into the wrong representation. This supports the idea that not all productive thinking is deliberate or conscious.

---

Creative and Flexible Approaches

These strategies emphasize generating novel solutions and breaking free from conventional thinking patterns.

Lateral Thinking

Coined by Edward de Bono, lateral thinking deliberately seeks unconventional angles. It rejects the assumption that straightforward logic alone leads to the best solutions.

One technique is provocation, where you make intentionally absurd statements to disrupt habitual thinking and open up new paths. Lateral thinking contrasts with vertical thinking, which digs deeper in one direction rather than exploring sideways. Vertical thinking refines an existing approach; lateral thinking abandons it for something unexpected.

Brainstorming

Brainstorming is a group technique that generates quantity over quality initially. The core rule is to suspend judgment so that even wild ideas get voiced. Collaboration can spark connections that individuals miss because different people bring different mental models.

Effectiveness depends on psychological safety. Research shows that criticism during the ideation phase kills creativity. People self-censor when they fear judgment, which defeats the entire purpose.

Analogical Problem Solving

This strategy transfers solutions from a source problem to a target problem. It requires recognizing that two problems share the same deep structure, even if they look completely different on the surface.

Gick and Holyoak's radiation problem is the classic demonstration:

1. Participants read a story about a general who divides his army into small groups that converge on a fortress from multiple directions.
2. They then had to solve a problem about destroying a tumor with radiation without damaging surrounding tissue.
3. The structural parallel is that you can divide the radiation into weaker beams aimed from multiple angles, just as the general divided his army.
4. Most participants didn't spontaneously see the analogy. They needed a hint to map the military strategy onto the medical scenario.

This shows that surface features often mislead. Deep structural mapping is cognitively demanding, which is why analogical transfer frequently fails without prompting.

---

Decomposition Strategies

These approaches manage complexity by breaking problems into smaller pieces. They reduce cognitive load by limiting how much must be held in working memory at once.

Divide and Conquer

Divide and conquer splits a large problem into independent sub-problems. You solve each one separately, then combine the solutions.

This reduces working memory demands because you only need to focus on one component at a time. It's most effective when sub-problems don't interact. If changing the solution to one part forces you to redo another part, this approach becomes much less efficient.

Problem Space Theory

This is Newell and Simon's foundational framework for understanding problem-solving. A problem consists of three elements:

• An initial state (where you start)
• A goal state (where you want to end up)
• A set of operators (legal moves that transform one state into another)

Solving a problem means searching the problem space, finding a path from the start state to the goal state through intermediate states. Every strategy in this guide can be understood as a different way of navigating a problem space.

This framework also explains why some problems are hard. Large problem spaces with many branching paths and dead ends require sophisticated search strategies. A problem with a small, constrained space (like tic-tac-toe) is easy; one with a vast space (like chess) demands heuristics to avoid exhaustive search.

---

Cognitive Barriers to Problem-Solving

Understanding what blocks solutions is just as important as knowing strategies. These phenomena explain why smart people get stuck.

Mental Set and Functional Fixedness

Mental set is the tendency to reuse strategies that worked before, even when they no longer apply. The Luchins water jug experiments demonstrated this: after solving several problems with a complex formula, participants kept using that formula even when a much simpler solution was available. Prior success made them blind to the easier path.

Functional fixedness specifically involves the inability to see objects beyond their typical use. In Duncker's candle problem, participants were given a candle, a box of tacks, and matches, and asked to attach the candle to the wall. Most struggled to realize that the box of tacks could serve as a shelf, not just a container. They were "fixed" on its usual function.

Both reduce cognitive flexibility. Prior experience becomes a liability rather than an asset when a novel problem requires a fresh approach.

---

Quick Reference Table

---

Self-Check Questions

1. Which two strategies both involve breaking problems into smaller parts, and what distinguishes how they do so?

2. A student solves a physics problem by recalling a similar problem from math class. Which strategy does this illustrate, and what cognitive process makes it difficult?

3. Compare and contrast algorithmic and heuristic approaches: When would you choose each, and what are the costs of each choice?

4. Why might an expert mechanic struggle more than a novice to find an unconventional use for a wrench? Which two barriers explain this?

5. An FRQ describes someone stuck on a puzzle who solves it immediately after taking a walk. Which two concepts from this guide explain what happened, and how do they work together?