A question changes the field of view
Knowledge tells us what already works. Curiosity helps us notice when the rules have begun to change. A good question opens more than one answer and leads beyond the first convenient source.
Instead of asking only “how can this be faster?”, ask “which part need not exist at all?” or “who sees this problem from a completely different angle?”.
A rhythm of small experiments
Curiosity without action remains inspiration. A small experiment gives it direction: one hypothesis, limited cost and a clear observation. The goal is not instant success but a rapid improvement in the quality of the next question.
That rhythm is resistant to fashionable shortcuts. It lets us use new tools without outsourcing the thinking process to them.
A weekly curiosity cycle
A lightweight framework for turning inspiration into a small, testable move.
- 01Observe
Record friction, anomaly or surprise.
- 02Question
Form several competing explanations.
- 03Hypothesis
Choose one predicted change.
- 04Experiment
Bound the cost, time and scope.
- 05Observe the result
Separate fact from interpretation.
- 06Revise
Change the next question, not just the answer.
Connecting distant points
The most interesting ideas often appear between disciplines. Design can teach strategy empathy, technology can teach creativity scale, and history can remind innovation that human needs repeat.
Curiosity becomes an operating system when we regularly gather different perspectives, test the connections and leave room to change our minds.
A question as a working hypothesis
In Mateusz’s approach, curiosity is not the random accumulation of interesting information. It begins with a question capable of changing a decision or a way of seeing the problem. Instead of asking only “how can this be faster?”, ask “which step need not exist?”, “which assumption are we treating as proven?” and “who experiences this problem differently?”. The question should open several possible answers while remaining concrete enough to guide a search for evidence.
Mateusz proposes recording why the question matters and what evidence could change the current view. That note reduces the temptation to collect only material supporting the first intuition. It also makes it possible to return later and judge whether the search produced new knowledge. Curiosity then becomes a practice for making better decisions rather than an endless collection of bookmarks, videos and tools that are never revisited.
A portfolio of sources instead of one bubble
One source rarely provides the entire picture. A vendor is well placed to describe product features but has a natural reason to emphasize success. A research paper explains method yet may not answer questions about daily use. A practitioner reveals real friction but represents one experience. Mateusz’s model combines a primary source, independent analysis, a user perspective and a critical view, then records where they agree and where their accounts diverge.
Diversity does not mean creating artificial balance between evidence and opinion. Sources should be weighted by proximity to the event, methodological quality, author incentives and verifiability. Curiosity requires openness to an unexpected argument, but it does not require treating every claim as equally credible. Looking beyond one industry, culture and familiar circle is valuable when the distinction between measurement, interpretation and speculation remains explicit.
From a signal to reusable knowledge
An experiment record should preserve the distinction between what was observed and how it is interpreted.
- 01Signal
What exactly drew attention?
- 02Alternatives
Which other explanations fit the observation?
- 03Prediction
What should happen if the hypothesis is right?
- 04Test
What is the smallest experiment that separates the explanations?
- 05Decision
What changes, what remains unknown and when will the question be revisited?
A small experiment turns inspiration into knowledge
Learning about an idea does not show whether it works in a particular setting. Mateusz proposes a small experiment with one hypothesis, limited cost, a short duration and an observation recorded in advance. Instead of redesigning an entire process, test one task, one week or a small group. The result need not confirm the idea. Its main purpose is to improve the next question and expose assumptions that reading alone did not reveal.
A sound experiment has a stopping condition and an easy route back. If it requires a major investment, irreversible publication or risk to other people, it is no longer a small test. Negative findings should also be recorded because they prevent the same failed attempt from returning under a new label. This rhythm makes it possible to use emerging tools without outsourcing the thinking process: technology accelerates observation while people still define the question and interpret the result.
Curiosity as a repeatable system
Mateusz’s operating rhythm contains four actions: collect questions, select one direction, run a small test and reflect briefly. Once a week, discard questions that are only momentary distractions and choose one that could affect a project, relationship, health decision or working method. After the experiment, record what changed, what remains unknown and which next step offers the strongest ratio of potential value to cost.
The system should connect distant fields. Sport can teach a technology team about feedback, design can bring empathy into strategy, and history can reveal recurring needs beneath new language. Connecting disciplines is not a matter of superficial analogy; it requires checking which mechanisms are genuinely shared. Curiosity becomes an advantage when it repeatedly moves from a different perspective to a testable idea and makes changing one’s mind a sign of learning rather than failure.
A question portfolio protects curiosity from accidental drift
A question list becomes useful when it reveals dependencies and the cost of neglect. Each portfolio entry records the decision it could affect, the horizon in which an answer remains useful and the condition that would make the subject obsolete. Questions resting on the same assumption are linked because different searches may be spending attention on one unknown. Priority goes to a question that unlocks others or protects a project from an expensive mistake, not merely to the most exciting topic.
A healthy portfolio balances questions close to current work with questions that probe whether the rules are changing. Exploratory topics need a stated reason for remaining visible so they do not become decorative curiosities. Review means changing priority, merging duplicates and deliberately parking work, not automatically adding entries. Curiosity then has capacity: every new question competes for attention and must show which decision, risk or possibility it helps the team see more clearly.
Deliberate dissent tests a claim before attachment hardens
Dissent is most valuable before a success narrative settles. For an important claim, someone constructs the strongest fair opposing case: an explanation a reasonable person could accept, not a caricature. The counter-brief names alternative causes, inconvenient evidence and a consequence expected if the rival account is true. Its purpose is to locate where confidence has moved ahead of evidence and turn disagreement into a testable distinction, not to win a debate.
Deliberate dissent needs protection from ritual. Rewarding objections alone creates theatrical pessimism, while letting the claim’s author choose every counterargument invites an easy opponent. Roles should be separated, the original critique preserved beside the response, and any resulting decision change recorded. A claim may emerge stronger, narrower or retired. Each outcome is useful because dissent serves reasoning rather than status, and a resolved objection leaves an auditable trace.
A hypothesis register preserves the history of learning
A hypothesis should be a versioned record, not a sentence silently repaired after every result. Its card separates observation, interpretation and prediction. It names the proposed mechanism, its expected trace, evidence capable of contradicting it and assumptions inherited from other questions. This reveals whether a team predicted an outcome or explained what it had already seen. Recording each change also reconstructs when knowledge entered the process.
The register is not a graveyard for failed ideas. A hypothesis may be narrowed, replaced or retired while its earlier form stays readable. Retirement includes the direct reason and a condition for reopening. This prevents the same idea returning under a new label and distinguishes a refuted claim from one never practicably tested. Failure memory becomes an asset: it marks the explored boundary and gives later work a better starting point than recollection.
A transfer test separates an analogy’s mechanism from its language
An analogy is productive when it transfers a causal relationship rather than attractive vocabulary. The test describes the source mechanism without metaphor and maps its elements to the target domain. It identifies relationships that must remain invariant, missing counterparts and the point where transfer should fail. If no breaking point can be named, the comparison is probably flexible enough to explain anything and precise enough to explain little.
The strongest transfer test is a new prediction: an observation or intervention that surface resemblance alone would not select. It is compared with a simpler account drawn only from the target domain. If both imply the same action, the analogy added inspiration rather than explanatory power. If they diverge, the difference creates a discriminating test. Cross-disciplinary thinking then becomes a controlled method with a clear route for discovering that the transfer was mistaken.
An evidence map exposes both the argument and its gaps
Notes arranged by source obscure which claim is supported. An evidence map reverses the order: concrete claims occupy the centre, while materials connect as support, contradiction, context or unresolved indication. Dependencies among claims are separate because strong evidence for one step cannot repair a weak earlier assumption. Empty nodes remain visible instead of disappearing beneath fluent summary, directing inquiry toward genuine gaps rather than more material for an already secure claim.
The map should not pretend to calculate confidence. It records each item’s provenance, directness, limitations and possible incentive conflict. The same source may strongly support an event description while weakly supporting a forecast. A decision view highlights nodes whose revision could reverse the plan, so inquiry follows the decisive gap rather than the most available content. The map becomes a shared interface between investigation, criticism and action while preserving the line between evidence and interpretation.
A synthesis rhythm creates decision memory, not a material archive
Synthesis need not follow every new reading. It belongs after an event that changes the map: contradictory evidence appears, a material gap closes, a constraint shifts or a decision approaches. A concise update states what is known, what changed, which hypothesis gained or lost support and what remains undecidable. The report becomes a history of the team’s evolving model rather than a summary of everything consumed, connecting information to consequence instead of rewarding accumulation.
Decision memory completes the practice. For a consequential choice, it preserves context, alternatives, reasons for rejection, accepted risk and the event that should trigger review. Its purpose is not to defend the past but to compare a decision fairly with knowledge available at the time. Without the record, teams remember more certainty than they possessed or judge earlier work with later evidence. Memory improves the reasoning process, not merely the final verdict.
A stopping rule turns inquiry into a decision
Inquiry can continue indefinitely because every answer opens another question. A stopping rule is set before fatigue or attachment writes it. It combines the decision’s nature, delay cost, reversibility and the chance that more evidence would change the choice. An easily reversed action can proceed with lower confidence; a difficult-to-undo outcome demands stronger support. Stopping is also legitimate when decisive uncertainty remains but no available method can resolve it.
Stopping does not mean declaring truth. The closing note separates the chosen action, current confidence, unresolved questions and signals for reopening. A team can act under incompleteness without pretending uncertainty disappeared. The rule guards against closing after the first persuasive account and using curiosity to avoid responsibility. Mature inquiry ends when more information is less valuable than a deliberate move paired with a prepared route for correction.
Frequently asked questions
Is “curiosity as a system” a scientific method?
No. It is Mateusz Więcek’s editorial framework for managing questions, sources, small tests and changes of mind. It does not replace domain expertise or an appropriate research methodology.
How do I choose a question worth pursuing?
Choose one whose answer could change a concrete decision. Record what is missing, which evidence could change the current view, the attention budget and the decision date.
How many sources are enough?
The framework sets no number. Provenance matters more: determine whether materials provide independent evidence, represent relevant perspectives and expose disagreement, or merely repeat one original claim.
What should happen to a negative experiment?
Record its context, execution and limits as carefully as a success. Separate three possibilities: the hypothesis was weak, the test was not discriminating, or the potential benefit was smaller than the cost.