Summary of "Thinking, Fast and Slow"

Core Idea

Your brain operates two thinking systems: System 1 (fast, intuitive, automatic) and System 2 (slow, deliberate, effortful)
System 1 is efficient but systematically biased; System 2 is accurate but lazy—and you rarely engage it when you should
Most decision failures stem from trusting intuition on complex problems where it fails predictably

Anchoring: First numbers you see (even random ones) disproportionately influence estimates—consciously reject anchors and generate baselines independently
Availability heuristic: Judge probability by how easily examples come to mind—vivid/recent events feel more likely than they are; deliberately seek base rates
Representativeness: Believe detailed scenarios are more probable than simple ones (false); detail actually reduces probability; ignore stereotypes, weight evidence by sample size
Overconfidence: Underestimate uncertainty and overweight small-sample results; regression to the mean is real, not magical recovery
Hindsight bias: Reconstruct past beliefs as "obvious"—judge decisions by process quality, not outcomes; document predictions beforehand

For High-Stakes Choices:

Force System 2 engagement—slow down, question intuitions, demand contradictory evidence before deciding
Apply regression correction: Start with baseline average, adjust only 30-60% toward your intuitive prediction (feels wrong, works better)
Use reference-class forecasting: How often do similar projects succeed?—don't rely on optimistic projections
Conduct premortems: Imagine failure, work backward to identify blind spots before committing

Protect Against Environmental Manipulation:

Audit your environment—you're unconsciously primed by irrelevant words, images, and framing (money primes selfishness; age primes slow movement)
Reframe problems multiple ways before deciding; if your choice flips, you're being manipulated by framing, not logic
Watch for framing effects: "Lives saved" vs. "lives lost" reverses preferences—same math, different words

Value & Loss Psychology:

Losses hurt ~2x more than equivalent gains feel good (loss aversion)—frame costs as "investments" to reduce psychological pain
You value what you own ~2x more than equivalent items you don't (endowment effect)—expect 50% less when selling
Overweight certainty and rare events (insurance, lotteries)—recognize you're paying for emotional relief, not expected value

Implement mandatory checklists and decision routines across problem-framing, information-gathering, and review stages
Demand trusted critics review your decisions—observers spot errors actors can't see while cognitively busy
Create a decision vocabulary ("anchoring situation," "planning fallacy") to catch biases faster
Train staff in efficient meetings and decision processes—this is surprisingly underdeveloped

Next decision: Identify which bias threatens it (anchoring? planning fallacy? framing?); consciously counter it with formal procedure
Build a checklist: Three stages—how you frame the problem, what evidence you gather, how you review before committing
Document predictions beforehand to prevent hindsight bias from distorting your learning
On major choices: Conduct a premortem (imagine failure) and use reference-class forecasting instead of intuitive optimism
Create decision vocabulary: Label biases as they happen—speeds recognition and breaks their power