1. Two systems run the mind. System 1 is fast, automatic, and intuitive; System 2 is slow, effortful, and deliberate. Most behavior is driven by System 1, while System 2 lazily endorses its outputs unless something forces it to engage. Many cognitive errors come from System 1 producing an answer and System 2 failing to audit it.
2. Heuristics produce predictable biases. The mind substitutes hard questions ("How likely is this?") with easier ones ("How easily does an example come to mind?"). Availability, representativeness, and anchoring heuristics save effort but distort probability estimates in systematic ways. Knowing the bias doesn't eliminate it, but it lets you flag situations where you should slow down.
3. We are loss averse, not just risk averse. Prospect theory shows that losses feel roughly twice as painful as equivalent gains feel pleasurable. This asymmetry explains the endowment effect, status-quo bias, and why people refuse fair coin-flip gambles. Decisions about money, careers, and relationships are warped by where the "reference point" sits, not by absolute outcomes.
4. Framing changes everything. The same options described as "90% survival" versus "10% mortality" produce different choices, even from experts. Frames anchor the reference point and prime emotional responses before reasoning begins. Anyone who controls the frame controls the decision.
5. The experiencing self and the remembering self disagree. The experiencing self lives through every moment in real time, while the remembering self constructs a story dominated by peaks and endings. Memory, not lived experience, drives future choices — including which vacations we take and which jobs we keep. This explains why people repeat unpleasant experiences they "remember enjoying."
6. Confidence is a feeling, not evidence. Subjective certainty grows from the coherence of a story, not from how much information supports it. "What you see is all there is" (WYSIATI) — the mind builds confident narratives from whatever scraps are present and ignores what's missing. This is why experts in low-validity domains are routinely outperformed by simple statistical rules.
7. Regression to the mean is constantly misread. Extreme performances are usually followed by less extreme ones for purely statistical reasons, but we invent causal stories — praise hurt them, punishment helped them. Recognizing regression dissolves a huge class of management, sports, and parenting illusions. Most "what worked" stories are noise.
8. Slow thinking has limits but is your best tool. System 2 is effortful and easily depleted, so you can't run it on every decision. The practical move is to recognize the situations where intuition fails — low-validity environments, statistical reasoning, novel risks — and force deliberation there. Designing your environment and defaults matters more than trying to be smarter in the moment.
