Thinking, Fast and Slow

Miles: A bat and a ball cost one dollar and ten cents together. The bat costs one dollar more than the ball. How much does the ball cost?

Miles: Almost everyone's first answer is ten cents. It's wrong. If the ball were ten cents the bat would be a dollar ten and the total would be a dollar twenty. The ball costs five cents.

Miles: More than half of the students at several elite universities got that wrong. The interesting part isn't the error. It's that the wrong answer arrives instantly and feels correct, and the right one requires you to stop and check.

Miles: Today: Thinking, Fast and Slow, by Daniel Kahneman. In two sentences: the mind runs two systems, one fast and automatic and one slow and effortful, and the fast one is generating most of your judgments. It's very good, and it's wrong in predictable ways you can't introspect your way out of.

Miles: Sixty Books. Presented by Stoar Consulting and Coaching.

Miles: Sixty Books, day twenty six, Miles with you. The arrangement doesn't change: one book a day, sixty in all, out of the last hundred years of management writing. Each of them gets read twice over, once for leaders and teams who think differently, and once for a job that now has AI doing a share of it. Nora's arriving with the research.

Miles: Nora, yesterday ended on a judgment that felt like an assessment and wasn't one. Today we get the machinery behind that.

Nora: We do, and it's the most decorated book on the shelf. Its author won a Nobel prize in economics, which is remarkable for a psychologist who never took a course in economics. He's also the only author in this run who published a correction of his own chapter, in public, and that's the most candid thing I've read on this whole shelf.

Miles: Then we'll get to the correction, because it changes how you read the rest.

Miles: Two systems.

Nora: System one is fast, automatic, associative, and effortless. It reads emotion on a face, completes familiar phrases, drives on an empty road. System two is slow, deliberate, effortful, and lazy. It does multiplication, fills in tax forms, and checks system one's work when it can be bothered.

Miles: Huh.

Miles: And he's clear these aren't real brain parts.

Nora: Explicitly. They're a narrative device for describing two modes of processing, and he says so in the book. That hasn't stopped a decade of people drawing them as boxes.

Miles: The biases.

Nora: Dozens, and the useful ones for a manager are a handful. Anchoring, where an initial number contaminates a later estimate even when it's obviously arbitrary. Availability, where you judge probability by how easily examples come to mind, which is why people fear the vivid and ignore the common.

Miles: Loss aversion.

Nora: From prospect theory, the work that won the prize, developed with Amos Tversky. Losses loom larger than equivalent gains, roughly by a factor of two. Which explains why organizations defend existing arrangements far harder than they pursue better ones.

Miles: And the substitution idea.

Nora: The one I'd keep if I kept nothing else. Faced with a hard question, the mind quietly swaps in an easier one and answers that instead, without noticing. How likely is this project to succeed becomes how confident does the person presenting it seem. You've answered a question. It wasn't the question.

Miles: Right.

Miles: What's WYSIATI?

Nora: What you see is all there is. His name for the fact that system one builds a coherent story out of whatever information is present and doesn't register what's missing. Confidence comes from the coherence of the story, not from the amount of evidence, which is why people are most certain when they know least.

Miles: And the planning fallacy.

Nora: Systematic underestimation of time and cost, robust across domains and decades. His remedy is the outside view: instead of reasoning from your specific plan, find the distribution of outcomes for similar projects and start there.

Miles: And the man who wrote it.

Nora: Daniel Kahneman, born nineteen thirty four in Tel Aviv, raised partly in occupied France, where his family hid from deportation and his father died shortly before liberation. He served in the Israeli army as a psychologist, designing selection systems, and that work turns up in the book as one of his earliest encounters with the limits of expert judgment.

Miles: And the partnership.

Nora: With Amos Tversky, one of the most productive collaborations in the history of psychology. Tversky died in nineteen ninety six. Kahneman received the Nobel prize in two thousand two, and has been consistent that the work was joint. He died in twenty twenty four.

Miles: Two more ideas before the reckoning, because managers use both.

Nora: The remembering self and the experiencing self. His finding is that our memory of an episode is dominated by its peak and its ending rather than by its duration, which he demonstrated with genuinely uncomfortable experiments on pain. And that we make decisions using the remembering self, which is a poor historian.

Miles: And the application.

Nora: Endings carry disproportionate weight. A meeting, a project, a working relationship, gets remembered largely by how it finished. Which is worth a leader knowing on the day they run five minutes over and skip the close.

Miles: And the second idea.

Nora: The premortem, which he takes from Gary Klein. Before committing to a plan, assemble the team, state that it's a year later and the project has failed completely, and have everyone write down why. It works because it converts a prediction task, where nobody wants to be the pessimist, into an explanation task, where being specific is socially rewarded.

Miles: And that's reachable in a way most pushback isn't.

Nora: Written, independent, and everybody does it at the same moment, so raising a concern isn't an act of individual dissent that costs standing. It's the best device for getting a worry into the room that we've found in this whole series, and it takes ten minutes.

Miles: Sure.

Miles: Now the reckoning, and this is the unusual part.

Nora: The author told us himself. The chapter on priming, the research suggesting that subtle cues can substantially influence subsequent behaviour, was hit hard by the replication crisis. In twenty seventeen, Kahneman wrote publicly that he had placed too much faith in underpowered studies and that the chapter's evidence was weaker than he'd believed.

Miles: A Nobel laureate saying chapter four of my book overstated the case.

Nora: And he went further, saying he'd known all he needed to know, since his own early paper with Tversky warned against trusting results that were statistically significant but based on small samples. It's the most credible passage in the book, and it's a correction.

Miles: What else is contested?

Nora: Ego depletion, the idea that self control draws on a limited resource that gets used up, features in the book and has had serious replication difficulty. Some of the specific effect sizes across the behavioural economics literature have shrunk under scrutiny.

Miles: And what has survived?

Nora: The core is robust. Anchoring, availability, loss aversion, the planning fallacy, and overconfidence have all replicated widely. Prospect theory is a foundational result. The two systems framing is a metaphor and not a finding, and it's still a useful one.

Miles: And the practical limit.

Nora: On debiasing, his published position was blunt. He said repeatedly that knowing about these effects doesn't protect you from them, that he hadn't improved his own intuitive judgments, and that the realistic remedy is organizational rather than personal. Design the decision process; don't expect people to think their way out.

Miles: Which tier?

Nora: The most rigorously grounded book on this shelf, and it's still a popular book and not peer reviewed research. A chapter its own author disowned, and an unusual amount of plain speaking about what it doesn't solve.

Miles: Now the turn.

Nora: Start with substitution, because it's the mechanism behind almost every hiring and evaluation problem this series has described.

Miles: Go on.

Nora: The hard question is will this person do the job well. That requires evidence you don't have. So the mind swaps in an easier question that has an immediate answer. Did this conversation feel comfortable. Did they seem confident. Do they remind me of people who worked out.

Miles: And the answer feels like an assessment.

Nora: It feels like judgment, and it arrives with confidence, and the confidence comes from the story hanging together rather than from its accuracy. That's culture fit, described mechanically. And it's why the person who doesn't produce familiar social signals gets filtered out by an interviewer who'd sincerely deny doing any such thing.

Miles: We met that in episode nine.

Nora: The naturalness bias, where experts preferred the performer described as a natural while stating they valued hard work. Same structure. Stated criterion, substituted criterion, no awareness of the swap.

Miles: What does a leader do about it?

Nora: Kahneman's own answer, which isn't think harder. It's structure. Decide the criteria before you see the candidate. Score dimensions separately rather than forming a global impression. Use work samples, which put the real question back in place of the substituted one. He used a version of it in the army, scoring traits one by one from factual questions, and the sum beat the interviewers' impressions.

Miles: And that structure is also an accessibility measure.

Nora: Precisely, and neither he nor anyone in that literature would describe it that way. A structured interview with predefined criteria and a work sample is the single most effective change an organization can make for autistic candidates, and it was arrived at purely for accuracy.

Miles: What about the biases themselves, in our context?

Nora: Two things. First, availability. When a manager thinks about how a neurodivergent employee is performing, the instances that come to mind most easily are the vivid ones, which are usually the difficulties. A person can have a strong quarter and one memorable bad afternoon, and the afternoon dominates the impression.

Miles: And the fix is a record.

Nora: Written contemporaneous notes on outcomes, which is what episode three's time log did for time. Memory isn't the instrument you think it is.

Miles: And the second thing.

Nora: The planning fallacy, which lands with extra force on anyone with a variable relationship to time. Everybody underestimates. If your internal sense of duration is unreliable as well, the errors compound. His outside view remedy applies directly: stop estimating from the plan and start estimating from your own history of similar tasks, which requires having kept a record of it.

Miles: Is there a flip side?

Nora: One worth naming carefully. Some autistic cognition shows less susceptibility to certain framing effects in laboratory studies, meaning decisions stay more consistent when the same choice is put in different wordings. That's an intriguing finding rather than a settled one, and I wouldn't build a claim on it. But it's a reminder that the biases in this book aren't universal constants.

Miles: Now AI.

Nora: The obvious framing is that these tools are a system two prosthetic, and I think that's exactly half right.

Miles: The half that's right.

Nora: They're tireless where system two is lazy. Checking a number, running the outside view by asking how long projects like this usually take, generating the disconfirming case. That's effortful work humans skip, and the skipping is the whole problem.

Miles: And the half that's wrong.

Nora: These systems produce fluent, coherent, confident output, which is precisely the signature that gets system one to accept something. His what you see is all there is applies at full force: a well written answer that leaves out a crucial consideration doesn't feel incomplete. It feels finished.

Miles: So the risk isn't that it's wrong.

Nora: Yeah. The risk is that it's wrong in the most persuasive format available, and that the reader's fast system won't flag it, because coherence is the only signal that system has.

Miles: There's an anchoring problem in here too.

Nora: A serious one, and it's the bias in this book I'd worry about most in daily use. Ask for an estimate and a number comes back instantly. That number is now the anchor for every figure that follows, including yours, and the book's own finding is that an anchor works even when you know it was arbitrary. A first draft is an anchor with paragraphs around it.

Miles: So what protects you?

Nora: Write your own number down before you ask. It costs you a moment, and it's the difference between using the thing and being led around by it.

Miles: And the premortem, given all that.

Nora: Run it on the output, not just on the plan. It's a year from now and following this answer went badly, tell me why. That's the same conversion the premortem makes, a prediction turned into an explanation, and it's the one prompt I'd teach before any other.

Miles: And the practical rule.

Nora: Use it for the disconfirming half. Ask what would have to be true for this to be wrong, and what a competent critic would say. That's the job it does that human system two reliably refuses to do.

Miles: Anything else before we finish.

Nora: The fixed caution on all of this now runs after the close, word for word the same every time, so it doesn't rely on me remembering to say it.

Miles: Then give me the book's last word.

Nora: He said the remedy is organizational rather than personal, and that's the sentence to carry out of here. Put the check into the process instead of into the reader, because the fast system is never going to ask for it.

Miles: That's Thinking, Fast and Slow.

Miles: Next time: Think Again, by Adam Grant, and the argument that the most valuable skill left is a willingness to be wrong on purpose.

Nora: After a book whose author rewrote his own chapter, that's the right one to follow with. I'll see whether Grant does it himself.

Miles: Nora, that was worth the wait.

Nora: I'll have finished it by then.

Miles: Read it, or don't. We did.

Miles: The credit for Sixty Books goes to Stoar Consulting and Coaching. Logan Williamson is a Licensed Professional Counselor in Texas and Louisiana. Consider this episode education about a book, made for the people who run teams and the people on them. It coached nobody, it counseled nobody, and it passed no judgment on you.

Nora: Every claim we made has a citation waiting on the episode page, right there with the transcript. And tomorrow, book twenty seven.

Miles: The part after the music. Miles and Nora are both generated voices. Neither voice belongs to a person, neither one is licensed, and nothing said here was pointed at you. AI has a hand in writing each script; Logan Williamson, a Licensed Professional Counselor in Texas and Louisiana, at Stoar Consulting and Coaching, then edits it and puts his name to it.

Nora: And one note about the AI segment. The evidence that these tools help neurodivergent people at work is early. There are controlled trials now, but they measure things like conversation skill rather than how a job actually goes, and for work itself most of what exists is individual providers reporting success. So take it as promising and not proven. The tools also get rebuilt every few months, so whatever you set up today is built on something that moves.