Problem Solving¶

Version: v2 · 2026-09-28 · History
Prime #: 1565
Domain group: Social Sciences
Origin domain: Psychology & Behavioral Sciences
Subdomains: Cognitive Psychology, Problem Space Theory → Psychology & Behavioral Sciences
Also from: Cognitive Science, Mathematics
Aliases: Solution finding, Solving

Core Idea¶

Problem solving is the organized attempt to bridge a gap between a present state and a goal when the route is not immediately available. The solver represents what is given, what counts as success, and what blocks direct attainment; generates or retrieves possible operations; tests candidate paths against constraints and evidence; learns from failure; and validates an outcome. The process may be deliberate or partly automatic, individual or distributed, symbolic or embodied. Its identity is the obstacle-to-solution transformation, not any single heuristic.

A problem is not merely an unpleasant condition. It supplies a goal or adequacy criterion, a present state, and constraints under which an acceptable transformation is unknown, contested, or costly to identify. Solving can proceed by search, decomposition, analogy, simulation, proof, experimentation, diagnosis, negotiation, design, or iterative construction. Well-structured problems make states and operators explicit; ill-structured problems require framing and representation to be revised during the attempt. In both cases, a proposed answer becomes a solution only relative to the criteria and constraints that generated the problem.

The Prime is broader than optimization, inquiry, decision making, troubleshooting, or refinement. Optimization selects a best feasible alternative under an objective; problem solving may seek any adequate route. Inquiry primarily reduces uncertainty; a solver may already understand the cause but lack a feasible intervention. Decision making selects among alternatives; solving often has to create them. Troubleshooting localizes and corrects a fault in an existing system. Refinement improves a candidate already in hand. These are common solving modes, but the invariant is the controlled passage from obstructed goal to validated adequacy.

How would you explain it like I'm…

Getting Past the Stuck Part

You want the cookie jar, but it's on a high shelf and you can't reach. You think: could I use a stool? Ask someone? Try one idea, and if it doesn't work, try another, until you get there. Problem solving is finding a way to get what you want when the way isn't obvious.

Finding a Way Through

Problem Solving is finding a way to reach a goal when the path isn't obvious. First you figure out where you are, where you want to be, and what's blocking you. Then you come up with ideas, try them, learn from the ones that don't work, and check whether your answer really does what you needed. You can break a big problem into small ones, use something similar you solved before, or experiment. It counts as solved only if it meets the goal and follows the rules of the problem.

Bridging Obstacle to Goal

Problem solving is the organized effort to close a gap between where things are and a goal when the path isn't immediately available. The solver figures out what's given, what counts as success, and what's blocking the way; then generates options, tests them against constraints and evidence, learns from failures, and checks the result. Methods include searching, breaking the problem into pieces, using analogies, simulating, proving, experimenting, and designing. Well-structured problems (like a math puzzle) have clear states and moves; ill-structured ones (like designing a better city park) require rethinking the problem along the way. Problem solving is broader than optimization (finding the single best option) or decision-making (choosing among existing options) — a solver often has to invent options and only needs one that is good enough. An answer counts as a solution only if it meets the criteria that defined the problem.

Problem Solving is the organized attempt to bridge a gap between a present state and a goal when the route is not immediately available. A problem, as opposed to a mere unpleasant condition, supplies a goal or adequacy criterion, a present state, and constraints under which an acceptable transformation is unknown, contested, or costly to identify. The solver represents givens, success criteria, and obstacles; generates or retrieves operations; tests candidate paths against constraints and evidence; learns from failure; and validates the outcome. Strategies include search, decomposition, analogy, simulation, proof, experimentation, diagnosis, negotiation, design, and iterative construction, and the process can be deliberate or partly automatic, individual or distributed. Well-structured problems have explicit states and operators, while ill-structured ones require revising the framing and representation during the attempt. It subsumes but is broader than optimization (a best feasible choice versus any adequate route), inquiry (reducing uncertainty, even when the cause is known but no intervention is), decision making (choosing among given alternatives rather than creating them), troubleshooting (fault localization in an existing system), and refinement (improving a candidate already in hand). A proposed answer is a solution only relative to the criteria and constraints that defined the problem.

Structural Signature¶

Recurring features:

solver. an organism, person, team, institution, algorithm, or coupled system capable of transforming or representing states
present state. the relevant facts, resources, commitments, and uncertainties at the start of the attempt
goal or adequacy criterion. the condition by which the problem is considered resolved
obstacle or gap. the missing route, conflict, unknown, constraint, or failure that blocks immediate attainment
problem representation. the chosen boundary, variables, relations, state encoding, and assumptions
operators. actions, inferences, experiments, constructions, negotiations, or transformations available to the solver
candidate generation. search, recall, analogy, decomposition, recombination, or invention of possible paths
testing and feedback. constraint checks, evidence, simulation, proof, measurement, or practical trials that eliminate or revise candidates
solution path. a transformation or sequence that bridges the represented gap
validation. confirmation that the result satisfies the criterion without violating binding constraints
learning loop. failed attempts can change strategy, representation, operator choice, or the criterion itself

What It Is Not¶

Problem framing. Framing establishes the boundary and success criteria; it is often a stage of solving but not the whole process.
Problem representation. Representation determines available states and operations; solving uses and may revise that representation to reach adequacy.
Decision making. Decision making chooses among alternatives, while problem solving may need to discover, construct, or test the alternatives first.
Optimization. Optimization seeks a best feasible result under a specified objective; many problems require only a satisfactory or demonstrably correct solution.
Inquiry. Inquiry seeks warranted understanding; solving may instead produce an intervention, proof, design, agreement, or repair.
Troubleshooting. Troubleshooting is problem solving specialized to diagnosing and correcting faults in an existing system.
Execution. Following a known procedure is not itself problem solving unless uncertainty or obstruction requires adaptation.

Broad Use¶

Problem Solving applies wherever its complete role structure is present; the source domain is historical provenance, not a boundary on the Prime.

Mathematics. a solver represents givens and constraints, applies transformations, and proves that a constructed answer satisfies them
Engineering design. teams generate architectures, simulate tradeoffs, test prototypes, and revise constraints
Diagnosis. evidence narrows competing explanations and selects an intervention whose response supplies further feedback
Software. developers reproduce a defect, localize causes, construct changes, and validate behavior with tests
Operations. organizations identify a performance gap, analyze constraints, trial countermeasures, and monitor recurrence
Science. an anomaly becomes a problem when existing models cannot jointly satisfy observation and theoretical constraint
Negotiation. parties search for arrangements that meet binding interests and institutional limits
Everyday action. people adapt routes, tools, and sequences when familiar procedures fail
Education. worked examples and feedback develop representations, heuristics, monitoring, and transfer
Artificial intelligence. search, planning, theorem proving, constraint satisfaction, and learned policies instantiate different solving mechanisms
Applicability boundary. A case must identify the carrier, criterion or shared objective, operative relation, and collapse condition defined above; verbal resemblance alone does not instantiate Problem Solving.

Clarity¶

Use Problem Solving only after naming the carrier and the relation that makes the case one. The minimal definition is: Problem solving is the goal-directed process of representing an obstructed or underdetermined state, generating or selecting possible transformations, testing them against constraints and evidence, and carrying an adequate path to a validated solution. The most common ambiguity is to substitute a neighbor, an enabling condition, or a consequence for the identity itself. A precise claim therefore fills at least these roles: solver, present state, goal or adequacy criterion, and learning loop. It also states the level of analysis and the evidence that would make a competent observer reject the classification. This prevents a familiar word from doing explanatory work that belongs to a testable relation.

Manages Complexity¶

Problem Solving manages complexity by compressing many implementations into a stable role structure without erasing the variables that control validity. Analysts can compare cases by filling the same slots, locate disagreement in a missing role, and separate the constitutive relation from causes, instruments, representations, and consequences. The compression is intentionally reversible: the short label expands into solver, present state, goal or adequacy criterion, obstacle or gap, problem representation, operators, with explicit boundaries for variation and collapse. This makes heterogeneous cases comparable while keeping local mechanisms, authority, uncertainty, and evidence visible.

Abstract Reasoning¶

Solver move. Identify an organism, person, team, institution, algorithm, or coupled system capable of transforming or representing states and record what observation supports the assignment.
Present State move. Identify the relevant facts, resources, commitments, and uncertainties at the start of the attempt and record what observation supports the assignment.
Goal Or Adequacy Criterion move. Identify the condition by which the problem is considered resolved and record what observation supports the assignment.
Obstacle Or Gap move. Identify the missing route, conflict, unknown, constraint, or failure that blocks immediate attainment and record what observation supports the assignment.
Problem Representation move. Identify the chosen boundary, variables, relations, state encoding, and assumptions and record what observation supports the assignment.
Operators move. Identify actions, inferences, experiments, constructions, negotiations, or transformations available to the solver and record what observation supports the assignment.
Candidate Generation move. Identify search, recall, analogy, decomposition, recombination, or invention of possible paths and record what observation supports the assignment.
Testing And Feedback move. Identify constraint checks, evidence, simulation, proof, measurement, or practical trials that eliminate or revise candidates and record what observation supports the assignment.
Counterexample move. Remove one constitutive role while holding the surrounding topic fixed; if the label still appears equally applicable, the definition has become too loose.
Neighbor move. Compare the filled roles with the distinctions below and route the case to the narrower or broader entry whose collapse test it actually satisfies.

Knowledge Transfer¶

Cross-domain transfer. The present-state–goal–obstacle–operator–test–solution relation is literally instantiated across cognitive, mathematical, engineering, organizational, and computational systems despite very different mechanisms. The invariant is not the source-domain vocabulary but the typed relation expressed by the one-line definition. Transfer is literal when a receiving domain can fill the same roles and expose the same failure boundary; it is analogical when only a visual, verbal, or narrative resemblance survives.

Local warrants for Problem Solving. Different domains establish those roles differently. Formal domains may use proof, specifications, or executable traces; empirical domains use observation and measurement; institutions use authority, records, and rules; practical settings use performance and counterfactual tests. This Prime coordinates the questions without replacing those standards.

Transfer procedure. First identify the source case without metaphor. Then abstract solver, present state, goal or adequacy criterion, obstacle or gap, problem representation. Map each role to the receiving case, test the collapse condition, and explicitly carry any domain qualification that affects validity. If a role has no literal occupant, transfer only a neighboring abstraction or label the comparison as analogy.

Examples¶

Formal/abstract¶

To prove that a graph has a requested property, a mathematician first converts the statement into explicit givens, constraints, and a success condition. Small examples expose invariants; a decomposition reduces the search; a candidate lemma is tested against counterexamples; and the final proof validates that every admissible graph satisfies the conclusion. The proof is a solution, while the sequence of representation, candidate generation, rejection, and validation is the solving process. Merely applying a memorized theorem after its applicability is obvious is routine execution at the boundary.

Mapped back: solver → the carrier in the example; present state → its stated counterpart; goal or adequacy criterion → the defining condition; learning loop → the described counterfactual boundary.

Applied/industry¶

A production service develops intermittent latency spikes. Engineers define the affected requests and acceptable latency, instrument the path, reproduce the condition, compare hypotheses, and discover that retries synchronize on a shared dependency. They test jitter and load-shedding changes, reject one because it violates durability, deploy the other gradually, and monitor recurrence. Diagnosis, experiment, design, decision, and validation all occur, but the higher-order identity is problem solving: an obstructed operational goal is transformed into a verified adequate state.

Mapped back: the applied carrier fills solver, its governing relation fills goal or adequacy criterion, the evidence fills operators, and removal of learning loop marks the point at which the example would cease to instantiate Problem Solving.

Structural Tensions¶

T1 — Representation versus reality. A tractable representation exposes operators but can omit the condition that actually controls success. Diagnostic: Which excluded variable or actor could invalidate every candidate solution?

T2 — Exploration versus exploitation. Generating diverse possibilities reduces fixation while testing and committing consume limited time and resources. Diagnostic: What evidence justifies narrowing the search now?

T3 — Decomposition versus interaction. Subproblems simplify reasoning, yet solutions can fail when independently solved parts interact. Diagnostic: Which constraints cross the proposed decomposition boundary?

T4 — Local adequacy versus system consequence. A fix can satisfy the immediate criterion while shifting cost, risk, or failure elsewhere. Diagnostic: Who or what lies outside the success metric but inside the causal system?

T5 — Persistence versus strategy change. Difficult problems require sustained effort, while repeating an unproductive representation wastes resources. Diagnostic: What failed prediction would trigger reframing rather than another attempt?

T6 — Solution versus learning. A one-off workaround may restore function without producing transferable understanding, while analysis may explain the issue without resolving it. Diagnostic: Does the task require immediate adequacy, reusable knowledge, or both?

Structural–Framed Character¶

Problem Solving is structural on the structural–framed spectrum. The present-state–goal–obstacle–operator–test–solution relation is literally instantiated across cognitive, mathematical, engineering, organizational, and computational systems despite very different mechanisms. Its structural skeleton is expressed by the roles above and can be evaluated without importing one home-domain mechanism.

Framing still matters for Problem Solving. Institutions choose some criteria, communities stabilize terminology, practices determine admissible evidence, and normative consequences can follow from classification. Those facts qualify an instance without making the whole abstraction conventional. The correct test is whether changing the frame removes the constitutive relation or merely changes how it is named, measured, governed, or valued.

Across the spectrum tests, the vocabulary travels with moderate qualification, evaluative weight is separable from the descriptive identity, institutional origin is not universally constitutive, and import must be distinguished from recognition. The entry is therefore portable but not context-free.

Substrate Independence¶

Problem Solving has high substrate independence — composite 5 / 5. The present-state–goal–obstacle–operator–test–solution relation is literally instantiated across cognitive, mathematical, engineering, organizational, and computational systems despite very different mechanisms. The carrier can change while the relation and its rejection test remain intact. A cross-domain use does not require one material, notation, institution, species, or implementation.

The Problem Solving rating is not a license for metaphor. Mechanisms and evidence remain local, and a familiar surface does not substitute for the typed roles. This Prime survives transfer only when the same relation is present and the same kind of counterfactual removal would make it fail.

Composite substrate independence — 5 / 5
Domain breadth — 5 / 5
Structural abstraction — 5 / 5
Transfer evidence — 5 / 5
Mechanism neutrality — 4 / 5

Relationships to Other Abstractions¶

Current abstraction Problem Solving Prime

Parents (1) — more general patterns this builds on

Problem Solving is part of Problem Representation Prime

Problem Solving contains Problem Representation because a solvable state, goal, operators, obstacles, and success test must be encoded before candidate transformations can be evaluated.

Children (1) — more specific cases that build on this

Incubation (Problem Solving) Domain-specific presupposes Problem Solving

A prepared problem-solving attempt is required before a break can count as incubation of that problem.

Hierarchy path (1) — routes to 1 parentless root

Problem Solving → Problem Representation → Representation → Abstraction

Neighborhood in Abstraction Space¶

Problem Solving sits in a sparse region of abstraction space (80^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely rather than landing on a neighbor.

Family — Failure, Robustness & Safety Margins (17 primes)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With¶

Problem framing. Framing establishes the boundary and success criteria; it is often a stage of solving but not the whole process.
Problem representation. Representation determines available states and operations; solving uses and may revise that representation to reach adequacy.
Decision making. Decision making chooses among alternatives, while problem solving may need to discover, construct, or test the alternatives first.
Optimization. Optimization seeks a best feasible result under a specified objective; many problems require only a satisfactory or demonstrably correct solution.
Inquiry. Inquiry seeks warranted understanding; solving may instead produce an intervention, proof, design, agreement, or repair.
Troubleshooting. Troubleshooting is problem solving specialized to diagnosing and correcting faults in an existing system.
Execution. Following a known procedure is not itself problem solving unless uncertainty or obstruction requires adaptation.

Solution Archetypes¶

No catalogued solution archetypes reference this prime yet.

Notes¶

No current live node supplies a defensible necessary genus for Problem Solving. Neighboring operations participate in some instances but do not entail the complete identity, so the node is admitted without a parent pending later graph densification.

The entry intentionally separates the abstraction from its causes, instruments, outcomes, moral appraisal, and common implementations. Future DAG or solution-archetype work may add non-hierarchical relations, but those links should be adjudicated independently rather than inferred from shared vocabulary.

References¶

Frozen Wikipedia revision: https://en.wikipedia.org/wiki/Problem_solving (revision 1369634964).
DOI: https://doi.org/10.4324/9781315806723
DOI: https://doi.org/10.1111/j.1467-8721.2007.00469.x
DOI: https://doi.org/10.1016/j.ijintrel.2011.12.009
DOI: https://doi.org/10.4135/9781446249215.n17
DOI: https://doi.org/10.1080/01650250143000319
DOI: https://doi.org/10.1093/geront/gnt118
DOI: https://doi.org/10.1080/09602010343000039
DOI: https://doi.org/10.1037/h0031360
DOI: https://doi.org/10.1016/j.acn.2003.08.006
DOI: https://doi.org/10.1037/a0025817
Supporting reference preserved in the packet: https://arxiv.org/pdf/2511.11738
Supporting reference preserved in the packet: https://www.researchgate.net/publication/238733375
Supporting reference preserved in the packet: https://books.google.com/books?id=K9Hm0UuFGJ0C&pg=PA189
Supporting reference preserved in the packet: https://www.researchgate.net/publication/247514323
Supporting reference preserved in the packet: https://www.tandfonline.com/doi/abs/10.1080/00221309.1932.9711880
Supporting reference preserved in the packet: https://web.archive.org/web/20200806135752/https://www.tandfonline.com/doi/abs/10.1080/00221309.1932.9711880
Supporting reference preserved in the packet: http://marketing.wharton.upenn.edu/ideas/pdf/armstrong2/DecompositionPrinciple.pdf
Supporting reference preserved in the packet: https://web.archive.org/web/20100620221713/http://marketing.wharton.upenn.edu/ideas/pdf/armstrong2/DecompositionPrinciple.pdf
Supporting reference preserved in the packet: https://www.doc.ic.ac.uk/~rak/papers/IFIP%2074.pdf
Supporting reference preserved in the packet: https://web.archive.org/web/20240119025430/https://www.doc.ic.ac.uk/~rak/papers/IFIP%2074.pdf
Supporting reference preserved in the packet: https://www.doc.ic.ac.uk/~rak/papers/LogicForProblemSolving.pdf
Supporting reference preserved in the packet: https://web.archive.org/web/20231102032823/https://www.doc.ic.ac.uk/~rak/papers/LogicForProblemSolving.pdf

The frozen Wikipedia revision is discovery provenance. The retained source set was reviewed for identity, formal or operational relation, and scope. The encyclopedia's structural synthesis is bounded to those claims; a thin authority surface is recorded as a nonblocking source-strengthening repair rather than concealed.