Critical Thinking & Problem-Solving

Lesson 1

1.1 How the Mind Forms Judgements

QUADRAEDGECritical Thinking & Problem-Solving

Module 1 · Lesson 1.1

How the Mind Forms Judgements

Most of your thinking is fast, automatic and invisible to you. Critical thinking is the deliberate decision to slow a small part of it down.

ThinkDiagnoseDecide

FastAutomatic, effortless, usually right, occasionally confidently wrong.

SlowDeliberate, effortful, and only available when you notice you need it.

TriggerThe skill is recognising which situations deserve the slow system.

Mechanism

Two ways of thinking

Psychologists commonly describe two modes. The fast mode recognises patterns, jumps to conclusions and runs constantly — it is how you read a colleague's mood or judge a familiar risk in a second. The slow mode reasons step by step, checks assumptions and costs effort. Fast thinking is not a defect; you could not function without it. The problem is that it does not announce when it is out of its depth.

Shortcuts

Heuristics and where they break

Fast thinking uses shortcuts: judge by what comes to mind easily, judge by resemblance, judge by the first number mentioned. These work well in familiar, stable situations with quick feedback. They fail in unfamiliar, complex or high-stakes situations — which is exactly where important decisions live.

Framing

The same fact, two answers

How a question is worded changes the answer. A treatment described as having a 90% survival rate is chosen more often than one described as having a 10% mortality rate, though they are identical. Before deciding, restate the problem in a second framing and check whether your preference survives.

When to deliberately slow down

High stakesThe cost of being wrong is large or hard to reverse.

UnfamiliarYou have little relevant experience to pattern-match against.

Strong feelingYou feel certain quickly, or you feel defensive.

Time pressureSomeone wants an answer now — often the reason the answer is bad.

Emotive framingThe question arrives loaded with words that do the deciding for you.

Key takeaway

Fast thinking is efficient and usually right. Critical thinking is knowing which few situations deserve the slow, deliberate mode — and having a phrase ready to buy the time.

Practical Exercise

Keep a two-day log of decisions you made in under ten seconds. Mark which ones were genuinely routine and which deserved more thought. Look for the pattern in the second group.

Summary

The mind runs on fast automatic judgement and slow deliberate reasoning. Critical thinking means recognising high-stakes, unfamiliar or emotionally loaded situations and switching modes on purpose.

Resources for this lesson

Web link

Lesson content: 1.1 How the Mind Forms Judgements

Open resource

Lesson 2

1.2 Recognising Bias in Yourself and Others

QUADRAEDGECritical Thinking & Problem-Solving

Module 1 · Lesson 1.2

Recognising Bias in Yourself and Others

Biases are not stupidity. They are systematic patterns in how everyone's thinking goes wrong — which means they are predictable, and predictable things can be designed around.

ThinkDiagnoseDecide

SystematicNot random errors — the same tilts, in the same directions.

InvisibleYou cannot feel a bias operating. You can only check for it.

DesignableProcess beats willpower: build the check into how you decide.

Catalogue

The ones that cost most at work

Confirmation bias makes you notice evidence that fits your view and dismiss the rest. Anchoring lets the first number mentioned drag every later estimate towards it. Availability makes vivid recent events feel more likely than common quiet ones. Sunk cost keeps you funding a failing project because of what you already spent. Overconfidence makes your estimate range too narrow, almost always.

Social

Bias in groups

Groupthink suppresses dissent in cohesive teams, especially when a senior person speaks first. The fundamental attribution error makes you explain other people's failures by their character and your own by your circumstances. Both are amplified by hierarchy and time pressure.

Countermeasures

What actually helps

Knowing about biases barely reduces them; changing the process does. Ask people to write their view before the meeting rather than after hearing the boss. Actively seek disconfirming evidence. Assign someone to argue the opposite case. Judge the next spending decision on future value only, ignoring what has already gone.

Five biases and their countermeasures

ConfirmationAsk: what evidence would change my mind? Then go looking for it.

AnchoringMake your own estimate before you hear anyone else's number.

AvailabilityAsk for base rates: how often does this actually happen?

Sunk costDecide as if you were joining today, with the money already gone.

GroupthinkCollect written views independently before the discussion opens.

Key takeaway

Biases are systematic and invisible from the inside, so the fix is procedural: independent views first, disconfirming evidence deliberately sought, past costs excluded.

Practical Exercise

Take a decision your team faces now. Write the strongest evidence against your preferred option, and name one process change that would surface such evidence routinely.

Summary

Confirmation, anchoring, availability, sunk cost, overconfidence and groupthink distort judgement predictably. Countermeasures are procedural rather than motivational.

Resources for this lesson

Web link

Lesson content: 1.2 Recognising Bias in Yourself and Others

Open resource

Lesson 3

1.3 Spotting Faulty Logic

QUADRAEDGECritical Thinking & Problem-Solving

Module 1 · Lesson 1.3

Spotting Faulty Logic

An argument can be persuasive and invalid at the same time. Learning a handful of recurring faults lets you hear the difference.

ThinkDiagnoseDecide

ClaimWhat is being asserted.

EvidenceWhat is offered in support.

InferenceThe step from one to the other — where most arguments fail.

Anatomy

Take the argument apart

Every argument has a claim, some evidence and an inference connecting them. Most weak arguments have acceptable evidence and a broken inference: the facts are true but they do not support the conclusion drawn. Asking 'does this follow?' separately from 'is this true?' catches more errors than either question alone.

Fallacies

The common faults

Attacking the person rather than the argument. Misrepresenting a position so it is easier to defeat. Presenting two options as if no third exists. Treating one example as proof of a pattern. Citing authority outside its field. Assuming that because one thing followed another, it was caused by it. Circular reasoning where the conclusion is quietly assumed in the premise.

Causation

The most expensive confusion

Correlation is not causation. Two things moving together may share a common cause, may be coincidence, or may run in the opposite direction to the one you assume. Before acting on a correlation, ask what else could produce it and whether the timing supports your explanation.

Six faults, and the question that exposes each

Ad hominemWould the argument still be wrong if someone else made it?

Straw manWould they recognise their position in the way it was just described?

False dilemmaIs there really no third option?

Hasty generalisationHow many cases is this based on, and were they typical?

Misplaced authorityIs this person an expert in this specific thing?

Post hocWhat else changed at the same time?

Key takeaway

Test 'is it true?' and 'does it follow?' as separate questions. Most persuasive-but-wrong arguments fail at the inference.

Practical Exercise

Take one argument you encountered this week. Write its claim, evidence and inference on three lines, then name the strongest objection to it.

Summary

Arguments consist of claim, evidence and inference. Common fallacies, and confusion of correlation with causation, usually break the inference rather than the facts.

Resources for this lesson

Web link

Lesson content: 1.3 Spotting Faulty Logic

Open resource

Lesson 4

2.1 Defining the Actual Problem

QUADRAEDGECritical Thinking & Problem-Solving

Module 2 · Lesson 2.1

Defining the Actual Problem

Most failed problem-solving is excellent work on the wrong problem. The definition stage is short, unglamorous and decides everything after it.

ThinkDiagnoseDecide

SymptomWhat you noticed. Usually where people start solving.

ProblemThe underlying gap between what is happening and what should be.

ScopeWhat is inside your control, and what is context.

Distinction

Symptoms are not problems

'Staff are not using the new system' is a symptom. The problem might be inadequate training, a workflow that is genuinely slower than the old one, missing permissions, or a manager who told people not to bother. Solving the symptom — more emails reminding people to use the system — is the most common form of wasted effort in organisations.

Statement

Write it down properly

A usable problem statement says what is happening, where and since when, how big it is, who is affected, and what should be happening instead. It contains no proposed solution. If your statement includes the word 'because' followed by a cause you have not verified, you have jumped ahead.

Boundaries

Decide what you are actually solving

Scope determines feasibility. Some factors you control, some you can influence, some you must simply accept and design around. Being explicit about which is which prevents both paralysis and the endless meeting about things nobody in the room can change.

SymptomWhat was noticed — the complaint, the number, the incident.

GapWhat is happening versus what should be happening, quantified.

ScopeInside our control, within our influence, or fixed context.

SuccessWhat must be observably true for this to be solved.

Key takeaway

Define the gap, quantify it, exclude solutions from the statement, and be explicit about what you control. Everything downstream depends on it.

Practical Exercise

Write a one-sentence problem statement for a real issue: what is happening, since when, how big, who is affected, what should be happening. Include no solution.

Summary

Effective problem-solving starts by separating symptoms from the underlying gap, writing a solution-free problem statement, and setting explicit scope.

Resources for this lesson

Web link

Lesson content: 2.1 Defining the Actual Problem

Open resource

Lesson 5

2.2 Finding Root Causes

QUADRAEDGECritical Thinking & Problem-Solving

Module 2 · Lesson 2.2

Finding Root Causes

Fixing what you can see gives you the same problem again next quarter. Root cause analysis is the discipline of asking one more question than feels necessary.

ThinkDiagnoseDecide

Ask whyEach answer becomes the next question, until you reach something fixable.

Look widePeople, process, tools, materials, environment, measurement.

Expect severalSerious problems rarely have exactly one cause.

Method

The five whys, done properly

Start with the problem and ask why it happened; then ask why of that answer, and keep going until you reach a cause you can actually change. Five is a rule of thumb, not a rule. Two conditions make it work: each answer must be evidenced rather than assumed, and the chain must stay honest — if the answer is 'because nobody checks', do not stop at 'staff were careless'.

Breadth

One chain is not enough

A single line of whys finds one path. Structured category checking — people, process, tools, materials, environment, measurement — surfaces contributing causes the chain missed. Most significant failures have a main cause and two or three contributors, and fixing only the main one leaves the system fragile.

Systems

Blame is not a root cause

'Human error' is where analysis stops being useful. If a person could make that mistake, the design permitted it — ask why the process allowed the error, why it was not caught, and why the consequence was severe. Systems questions produce fixes that survive staff turnover; blame produces a memo.

Problem: the monthly report went out with wrong figures.

Why? The source spreadsheet was not the latest version.

Why? Two versions exist and the file names look identical.

Why? There is no naming convention or single source of truth.

Why? Nobody owns the reporting data — fixable, and the real cause.

Key takeaway

Ask why until you reach something you can change, check breadth across categories, and treat 'human error' as the beginning of the analysis rather than its conclusion.

Practical Exercise

Apply the five whys to one recurring problem, then check for contributing causes across people, process, tools and environment. Mark which links are evidenced and which are assumed.

Summary

Root cause analysis uses iterative why questions plus structured category checking, treats 'human error' as a design question, and expects multiple contributing causes.

Resources for this lesson

Web link

Lesson content: 2.2 Finding Root Causes

Open resource

Lesson 6

2.3 Evaluating Information and Evidence

QUADRAEDGECritical Thinking & Problem-Solving

Module 2 · Lesson 2.3

Evaluating Information and Evidence

Everything now arrives with confidence attached. The useful question is not whether a claim sounds right, but what would have to be true for it to be right.

ThinkDiagnoseDecide

SourceWho produced this, how, and what do they gain?

TypeFact, inference or judgement — they need different tests.

StrengthSample, method, base rate, and who disagrees.

Provenance

Interrogate the source

Ask who produced the information, by what method, when, and with what interest in the result. Prefer primary sources over descriptions of them — a great deal of confident reporting turns out to summarise a summary. Funding and incentive do not make a claim false, but they tell you how hard to look.

Type

Sort claims before testing them

A fact can be verified. An inference is a conclusion drawn from facts and must be checked for logic. A judgement expresses a value and cannot be settled by evidence alone — it can only be argued. Most unproductive arguments mix all three and treat them as one.

Strength

The four questions that do most of the work

How many cases, and were they representative? Compared with what — what is the base rate or the control? What would we expect to see if this were false, and do we see it? And who credible disagrees, and on what grounds? A source that has never engaged a serious counter-argument has not been tested.

Testing a claim in four questions

SampleHow many, over what period, and were they typical?

ComparisonCompared with what? What is the base rate?

FalsificationWhat would we see if this were false — and do we?

DissentWho credible disagrees, and what is their strongest point?

Key takeaway

Check who produced it and why, sort facts from inferences and judgements, then ask about sample, comparison, falsification and credible dissent.

Practical Exercise

Take one claim relevant to your work. Apply the four questions and write a one-paragraph assessment of how much confidence it deserves.

Summary

Evaluating information means interrogating provenance, distinguishing facts from inferences and judgements, and testing sample, comparison, falsifiability and credible dissent.

Resources for this lesson

Web link

Lesson content: 2.3 Evaluating Information and Evidence

Open resource

Lesson 7

3.1 Generating Better Options

QUADRAEDGECritical Thinking & Problem-Solving

Module 3 · Lesson 3.1

Generating Better Options

Most decisions are made between the first two options anyone thought of. Widening the set costs an hour and changes more outcomes than any evaluation technique.

ThinkDiagnoseDecide

Diverge firstGenerate without judging. Judging early kills the useful oddities.

Beyond binary'Whether or not' questions are a symptom of a narrow option set.

Force varietyConstraints and opposites produce options you would not reach otherwise.

Separation

Two different activities

Generating options and evaluating them are separate tasks that fail when mixed. Evaluate while generating and people stop offering anything unconventional. Set an explicit rule: for the first fifteen minutes, nothing is criticised and nothing is committed to.

Widening

Prompts that produce real alternatives

What if the budget were half? What if it were triple? What would our best competitor do? What would we do if we could not do the obvious thing at all? What is the smallest version that would still count as progress? What if we did nothing — what actually happens? These questions reliably produce a third and fourth option, and the eventual choice is often a combination.

Combination

The best option is often a hybrid

Once you have five or six options, look for elements that combine: the speed of one with the reversibility of another. This is the practical payoff of divergence — not that option six wins, but that it contributes something to the option that does.

Key takeaway

Separate generating from evaluating, refuse binary framings, and use constraint prompts to force genuine alternatives before you judge any of them.

Practical Exercise

Take a live decision and generate six options using the widening prompts. Identify two that could be combined into a stronger seventh.

Summary

Better decisions start with wider option sets: generate before evaluating, reject binary framings, use constraint prompts, and look for hybrids.

Resources for this lesson

Web link

Lesson content: 3.1 Generating Better Options

Open resource

Lesson 8

3.2 Choosing with a Structured Method

QUADRAEDGECritical Thinking & Problem-Solving

Module 3 · Lesson 3.2

Choosing with a Structured Method

Structured decision methods rarely tell you something you could not have worked out. What they do is make your reasoning visible, comparable and arguable.

ThinkDiagnoseDecide

Criteria firstAgree what matters before you score anything.

Weight honestlyNot all criteria matter equally. Say so in advance.

ReversibilityCheap-to-undo decisions deserve less analysis, and faster action.

Matrix

Weighted criteria scoring

List your options as rows and your criteria as columns. Assign each criterion a weight reflecting how much it matters. Score each option against each criterion, multiply by the weight, and total. The number is not the decision — its value is that disagreements become specific: people can argue about a weight or a score rather than about a vague overall preference.

Doors

Match the method to the stakes

Some decisions are easily reversed; others are not. Reversible decisions should be made quickly by whoever is closest, because the cost of being wrong is small and the information from trying is real. Irreversible, expensive or reputationally significant decisions deserve the full method, more evidence and more people. Applying heavy process to light decisions is its own failure.

Pre-mortem

Imagine it already failed

Before committing, ask the group to assume it is a year later and the decision went badly, then write down why. This licenses doubts that hierarchy usually suppresses, and it surfaces risks in fifteen minutes that no risk register produced. Then decide what you would change to prevent the most plausible failure.

Option	Cost (w 3)	Speed (w 2)	Risk (w 5)	Weighted total
A — build in-house	2	2	4	30
B — buy off the shelf	3	5	3	34
C — pilot then decide	4	3	5	43
Scores 1–5. The output is a conversation, not a verdict.				C

Key takeaway

Make the reasoning visible: criteria and weights agreed in advance, effort matched to reversibility, and a pre-mortem before committing.

Practical Exercise

Build a weighted matrix for a real decision with at least three options and four criteria. Run a fifteen-minute pre-mortem on the leading option and note what you would change.

Summary

Weighted criteria matrices make reasoning explicit and disagreements specific; effort should scale with reversibility, and a pre-mortem surfaces risks before commitment.

Resources for this lesson

Web link

Lesson content: 3.2 Choosing with a Structured Method

Open resource

Lesson 9

3.3 Implementing, Testing and Reviewing

QUADRAEDGECritical Thinking & Problem-Solving

Module 3 · Lesson 3.3

Implementing, Testing and Reviewing

A decision is a hypothesis about the future. Implementation is the experiment, and reviewing it honestly is how thinking actually improves.

ThinkDiagnoseDecide

Test smallWhere possible, buy information before buying commitment.

Assign clearlyOwner, date, first step, and how you will know it is working.

Review processJudge the decision by how it was made, not only by how it turned out.

Piloting

Reduce the cost of being wrong

Where a decision can be tested at small scale — one team, one region, one month — the pilot buys information at a fraction of the cost of full commitment. Define in advance what result would count as success and what would make you stop, because deciding that afterwards is where sunk cost creeps back in.

Execution

The plain mechanics

Every action needs an owner, a date and a defined first step. Add leading indicators — early signals that things are on track — rather than waiting for the lagging outcome that arrives too late to act on. Name in advance what would trigger a review.

Learning

Separate the decision from the outcome

Good decisions sometimes produce bad outcomes and bad decisions sometimes get lucky. Judging only by results teaches your organisation the wrong lessons and rewards recklessness that happened to work. Ask instead: given what we knew and could have known, was this well reasoned? A short decision journal — the choice, the reasoning, the expectation, the date — makes this possible, because memory quietly rewrites what you predicted.

Good decision, good outcomeDeserved. Record what worked and why.

Good decision, bad outcomeBad luck or unknowable risk. Do not punish the reasoning.

Bad decision, good outcomeThe dangerous one — it teaches the wrong habit.

Bad decision, bad outcomeThe learning case. What was knowable at the time?

Key takeaway

Test small where you can, assign owner and date, watch leading indicators, and review the quality of the reasoning separately from the luck of the result.

Practical Exercise

Start a decision journal with one current decision: the choice, the reasoning, your expectation and a review date. Add two more this month.

Summary

Implementation means small tests where possible, clear ownership and leading indicators, and review that separates decision quality from outcome luck.

Resources for this lesson

Web link

Lesson content: 3.3 Implementing, Testing and Reviewing

Open resource