Insight Editions · No. 21
Emotional
How Feelings Shape Our Thinking
Leonard Mlodinow
Leonard Mlodinow examines how emotions arise from bodily states, experience, and culture, and how they shape attention, motivation, judgment, and everyday thought.
A study guide to Emotional by Leonard Mlodinow. Shared free for personal reading; not for sale and not endorsed by the authors or publishers. Please read and buy the original.
Introduction
Mlodinow opens with his mother, a Holocaust survivor, who answered a child's mess with "I wish I were dead! Why did I survive?" and warned him at dinner that he might one day "crawl on your belly through the mud" for filthy water. He grasped she was an outlier only when a psychiatrist, assessing her claim for German reparations, judged her traumatised. His father, a resistance fighter and Buchenwald survivor, went through comparable horror yet stayed optimistic and self-confident. Why do two people respond so differently to the same history? What are emotions, why do we have them, and how can we manage them?
The emotion revolution
The traditional theory, traced back to Darwin, held that:
- there is a small set of basic emotions (fear, anger, sadness, disgust, happiness, surprise), universal and distinct;
- each is triggered by specific stimuli and causes fixed behaviours;
- each arises in a dedicated brain structure.
It shaped Freud and the emotional intelligence idea of John Mayer and Peter Salovey, popularised by Daniel Goleman's 1995 book. Mlodinow's verdict: "But it is wrong." New tools (the connectome, optogenetics, neuroimaging) have created affective neuroscience. Basic emotions turn out to be categories of varied feelings; depression looks like a syndrome of subtypes; a phone app helps about a quarter of depressed users; scans can sometimes predict who benefits from psychotherapy rather than drugs.
We cannot think or decide without emotion, and emotions far more often lead us right than wrong. The brain is a prediction machine: the unconscious constantly applies past experience to forecast the consequences of the present. Emotion is part of that computation, not interference with it.
The plan: Part I asks what emotion is (evolution, purpose, core affect, mind and body). Part II covers pleasure, motivation and determination. Part III covers emotional profiles (with questionnaires) and emotion regulation. His mother, at 97, has "mellowed, but at her core she never changed."
About the author and approach
Leonard Mlodinow trained as a theoretical physicist (PhD, Berkeley), worked at Caltech, co-wrote A Briefer History of Time and The Grand Design with Stephen Hawking, wrote for Star Trek: The Next Generation, and wrote earlier popular books on randomness (The Drunkard's Walk), the unconscious (Subliminal) and flexible thinking (Elastic). Emotional (2022) completes a loose trilogy about the hidden workings of the mind. His method is to interview leading researchers (Adolphs, Anderson, Barrett, Berridge, Parvizi, Fenton-O'Creevy and others), retell landmark experiments, and weave in his family's history. The book is written for general readers; it has no equations and few technical terms, and it closes chapters with practical advice and, in Chapter 8, self-assessment questionnaires.
Why two survivors, two emotional lives
The introduction is not a warm-up; it is the research question of the whole book. Mlodinow's mother treated ordinary childhood messes as rehearsals for catastrophe. She had survived the camps and, decades later in Chicago, still spoke as if the next day might demand crawling through mud for water. A German-reparations psychiatrist confirmed what the family already felt: her trauma had locked an anxious emotional profile in place. His father, a resistance fighter who endured Buchenwald, emerged with a different set of defaults: optimism, self-confidence, a readiness to move on. Same historical horror, opposite emotional residues. The book asks what that difference is made of: genes, early care, luck of temperament, the stories people tell themselves, or some mix that science can now begin to name.
- The traditional theory (a small set of universal basic emotions, each with a dedicated brain structure and a fixed behaviour) shaped Freud, Darwin's popular reception, and Goleman's emotional intelligence, and it is wrong.
- New tools (connectome mapping, optogenetics, high-resolution imaging) created affective neuroscience, which treats emotions as constructed, flexible modes of mental operation.
- We cannot think or decide without emotion; emotions far more often lead us right than wrong.
What the traditional theory claimed
Darwin's The Expression of the Emotions in Man and Animals (1872) gave the traditional picture its scientific pedigree: a few universal expressions inherited from animals. Later writers hardened that into a tidy package: fear, anger, sadness, disgust, happiness, surprise; each triggered by specific stimuli; each housed in a dedicated structure (often the "limbic system"). The package travelled through Freud's hydraulic passions and into the 1990s classroom as emotional intelligence (Salovey and Mayer, then Goleman's 1995 bestseller). Mlodinow's point is not that those authors were fools; it is that the package does not survive contact with modern data. "Basic emotions" turn out to be categories covering many varied feelings. Depression is not one disease but a family of subtypes. A phone app helps about a quarter of depressed users; brain scans can sometimes predict who will benefit from psychotherapy rather than drugs. The old map is too coarse for the territory.
The prediction-machine thesis
The book's master metaphor is the brain as a prediction machine. The unconscious constantly applies past experience to forecast what the present will become. Emotion is not interference with that forecast; it is part of the computation that produces it. Valence and arousal from the body, context from the situation, and concepts from culture combine into a felt state that sets goals, directs attention, and re-weights evidence. That claim runs through every later chapter: bees made pessimistic by shaking; judges hungrier than they think; traders who reappraise losses; a Soviet officer whose gut feeling may have prevented nuclear war.
Structure of the book
Part I (Chapters 1 to 3) asks what emotion is: history of the concept, evolutionary purpose, core affect and the mind-body loop. Part II (Chapters 4 to 7) asks how emotion guides thought and motivation: modes of operation, constructed feelings, wanting versus liking, determination. Part III (Chapters 8 and 9) asks how to know and manage your own profile: inventories, nature and nurture, acceptance, reappraisal and expression. The Epilogue returns to his mother, who dies in the 2020 pandemic, and tests whether the book's tools survive when the author himself is overwhelmed.
His mother, at 97 when he wrote, had "mellowed, but at her core she never changed." Profiles are sticky. That stickiness is why Chapter 8's inventories and Chapter 9's regulation strategies matter: you cannot rewrite your childhood, but you can learn the shape of the system you are stuck with and work with it rather than against it.
TWO SURVIVORS → ONE QUESTION. Same trauma, opposite emotional profiles. The book is the search for what makes that difference, and for tools that work with profiles rather than against them.
Write two short paragraphs: one for a person you know who "never got over" a hard event, and one for someone who absorbed a similar blow and stayed open. Do not diagnose. Ask only: what differs in their default valence, their arousal, their stories about the future? That is the introduction's research question applied to your own life.
The survivor contrast can harden into a moral ranking: the optimistic father as hero, the anxious mother as defect. The book resists that. An anxious profile can save a life (the mole that was melanoma); an optimistic one can miss a threat. Profiles are tools with costs. Shadow: using the introduction to score people as "resilient" or "broken" inverts the point.
Two survivors, two profiles; traditional basic-emotion theory is wrong; emotion is a tool of the prediction machine. The book will show what emotion is, how it drives us, and how to know and manage your own profile.
1. Thought Versus Feeling
The Enterprise
On Halloween 2014 Virgin Galactic's SpaceShipTwo (VSS Enterprise) was dropped from its carrier at about 50,000 feet over the Mojave on its 35th test flight. More than 700 tickets had been sold at $200,000 to $250,000. Just before Mach 1, co-pilot Michael Alsbury unlocked the feathering air brake about 14 seconds early. Aerodynamic forces deployed it, and at 920 mph the ship tore apart. Alsbury died; pilot Peter Siebold was thrown out at supersonic speed and survived with no memory of it.
The NTSB blamed Scaled Composites (Northrop Grumman) for lacking a fail-safe against a single human slip; NASA-style programmes demand two-failure tolerance. It also suggested Alsbury, short of recent flight experience, was under high stress from the vibration and loads.
Researchers strapped honeybees into harnesses and shook them for 60 seconds. Trained that one odour meant sugar and another bitter quinine, the bees were then offered 2:1 mixtures, ambiguous smells. Shaken bees refused the mostly sweet mix more often than controls: a pessimistic cognitive bias, with lower dopamine and serotonin, as in anxious humans. The authors suggested "honeybees could be regarded as exhibiting emotions." The point: anxiety does not cause "mistakes"; it re-weights ambiguous evidence toward threat, sensible in a dangerous world.
Emotions affect thinking, usually for the better. A friend who lost his health insurance grew anxious about every symptom and so had a mole checked: it was melanoma, caught in time. Thirty-four successful Enterprise test flights made no news.
From Plato to Darwin
| When | Who | Idea |
|---|---|---|
| c. 370 BC | Plato (Phaedrus) | Soul as a charioteer (reason) driving a dark horse (appetite) and a noble horse (spirit) |
| Middle Ages | Christian philosophers | Passions as sins; love and compassion as virtues |
| 1660s | Thomas Willis | Names brain structures; "emotion" from Latin movere, to move |
| 1820 | Thomas Brown, Edinburgh | First modern use of "emotion"; lectures ran to 20 editions |
| 1872 | Darwin, The Expression of the Emotions | Universal expressions inherited from animals |
| 1960s to 2010 | MacLean, Sagan, Goleman | Triune brain: reptilian, limbic, neocortex |
Brown collapsed while lecturing in December 1819 and died on 2 April 1820, aged 42. Darwin collected questionnaires from missionaries, photographs of actors and babies, and notes on his son William. He concluded that emotions came from animal ancestors, which was revolutionary, but also that a "god-like intellect" had made them vestigial, like the appendix.
Male ruddy ducks grow chestnut plumage, a blue bill and a penis that can match their body length, all shed each year. Smaller males, anxious about bullying, moult sooner and grow far smaller organs, ceding mating and avoiding fights. Nature: "Sexual Competition Among Ducks Wreaks Havoc on Penis Size." Anxiety here serves an evolutionary function. So does parental love, which keeps us feeding helpless big-brained babies at 3 a.m.
The triune model (Paul MacLean, popularised by Carl Sagan's The Dragons of Eden and by Goleman) is "at best a vast oversimplification": the layers communicate constantly, and the neocortex helps create emotion, not merely restrain it.
Saved by emotion
On 1 September 1983 a Soviet Su-15 pilot shot down Korean Air Lines Flight 007, which had strayed into Soviet airspace, killing all 269 aboard, including Congressman Larry McDonald. "I saw two rows of windows and knew that this was a Boeing... But for me this meant nothing." The pilot followed procedure. Tensions soared; Andropov feared a US first strike.
In the Serpukhov-15 early-warning bunker, engineer-officer Stanislav Petrov (44) watched the screen flash "LAUNCH," then report five Minuteman missiles. The system rated reliability "highest" after 30 layers of verification; protocol said report. Something felt wrong, and he reported a false alarm, rating his own chance of being right at fifty-fifty. After 20 minutes nothing had landed. The cause: sunlight glinting off high clouds, read by satellites as launches. His gut feeling, drawing instantly on experience, beat the logical rule.
The Enterprise, stress and single-point failure
The VSS Enterprise crash on Halloween 2014 is not a parable about "feelings bad, procedures good." Co-pilot Michael Alsbury unlocked the feathering air brake about 14 seconds early, under vibration and load, short of recent flight experience. The NTSB blamed Scaled Composites for lacking a fail-safe against a single human slip; NASA-style programmes demand two-failure tolerance. Emotion here is the stress that narrows attention and rushes a practised action. The chapter's claim is subtler than "don't get stressed": emotion re-weights what counts as evidence, and design must assume that re-weighting will happen.
Thirty-four prior successful flights made no news. That is the base-rate trap that runs through the book: calm success is invisible; one anxious miss is a headline. The friend who lost health insurance and had a mole checked (melanoma, caught in time) is the inverse case: anxiety that looked like overreaction and was adaptive.
Plato's charioteer to MacLean's stack
The history section is load-bearing. Plato (Phaedrus) gave the West its durable metaphor: reason as charioteer, appetite as dark horse, spirit as noble horse. Medieval Christian writers sorted passions into sins and virtues. Thomas Willis in the 1660s named brain structures and helped pull "emotion" (from Latin movere, to move) into anatomy. Thomas Brown of Edinburgh, in lectures that ran to twenty editions after his death in 1820, gave the word its modern sense. Darwin (1872) made expressions universal and inherited, but also treated them as vestigial, like the appendix: a "god-like intellect" had outgrown them. Paul MacLean's triune brain (reptilian, limbic, neocortex), popularised by Carl Sagan and by Goleman, gave the old hierarchy an anatomy that textbooks and TED talks still recycle. The new science's first job is demolition: the layers communicate constantly; the neocortex helps create emotion, not merely restrain it; there is no emotion organ sitting below a reason organ.
| Claim | Status, today |
|---|---|
| Three stacked brains from evolution | Neuromyth; vertebrate brains are not layered that way |
| Limbic system = emotion | The "limbic" label covers regions that also do memory, smell, homeostasis |
| Neocortex restrains emotion | Cortex participates in constructing emotion |
| Popular use in EI training | Still common; treat as metaphor, not anatomy |
Bees, ducks and the function of anxiety
The shaken-bee study (Bateson et al., Current Biology, 2011) trained bees that one odour meant sugar and another bitter quinine, then offered ambiguous 2:1 mixes. Bees shaken for 60 seconds refused the mostly sweet mix more often; dopamine and serotonin fell. Anxiety did not cause "mistakes"; it re-weighted ambiguous evidence toward threat. Whether bees feel remains disputed; that they show judgement bias is not. Male ruddy ducks that anticipate bullying moult sooner and grow far smaller organs, ceding mating to avoid fights: anxiety with an evolutionary function. Parental love keeps big-brained infants fed at 3 a.m. when a cold cost-benefit ledger would walk away. Emotion, in this framing, is not noise on the channel of reason; it is a channel that evolved because fixed scripts fail in variable worlds.
Petrov: gut, logic and the A closer look
Stanislav Petrov on 26 September 1983 saw "LAUNCH" and five Minuteman missiles on a system rated highest reliability. Protocol said report. He reported a false alarm. Cause: sunlight on high clouds. The popular retelling says emotion beat logic. Petrov's own later account mixed gut unease with inference: a real first strike would use hundreds of missiles, and ground radar showed nothing. The lesson for decision systems (including trading floors and incident response) is not "trust your gut" or "trust the dashboard"; it is that experience-shaped affect and base-rate reasoning can correct each other when either alone would fail.
ENTERPRISE → BEES → PETROV. Stress re-weights evidence; anxiety can be adaptive; gut plus inference beat a rigid rule. Emotion is a tool of thought, not its enemy.
"Emotion is a tool of thought" can become a blank cheque for every impulse. Incidental moods (anger from traffic, anxiety from caffeine) still bias risk judgements. Petrov's night is not a licence to override every alarm. Shadow: treating every strong feeling as deep wisdom, or every procedure as stupid, both invert the chapter.
Ambiguous nights: expanding the cases
The chapter's three laboratory-of-life cases (Enterprise, bees, Petrov) share a statistical structure that is easy to miss when the stories are told for drama. Each involves a low-base-rate event under high arousal. Feather unlocks that early are rare; mostly-sweet odour mixes are usually safe; five-missile "launches" on a quiet night are more often glints than wars. Emotion's evolutionary job is to shift the decision threshold when the cost of a miss is asymmetric. Anxiety lowers the threshold for calling threat; calm raises it. Neither setting is "rational" in the abstract; each is a bet about which error is worse.
Enterprise as systems failure
Scaled Composites had flown dozens of captive and glide tests. Commercial tickets at $200,000 to $250,000 created public momentum. The feathering system was a known critical control; unlocking it in the wrong speed regime was unsurvivable. Two-failure tolerance would have required a second crew confirmation, an inhibit tied to Mach number, or both. High stress from vibration is not an excuse in the NTSB sense; it is a design input. Chapter 1 thus links affective science to engineering: if emotion re-weights attention, safety systems must assume re-weighting.
Bees and the politics of "emotion"
Calling bee states "emotion" upsets philosophers and some biologists. The chapter's careful move is to separate functional affective states (persistent, valenced, behaviour-guiding) from the full human phenomenology of feeling. Bateson et al. (2011) earned the functional claim. Solvi, Chittka and later bumblebee play and optimism work (2016 onward) strengthened it. The 2024 New York Declaration on Animal Consciousness treating insect sentience as a realistic possibility is a social marker, not a proof.
Petrov without the myth
Serpukhov-15 on 26 September 1983 sat inside a crisis month. KAL 007 had been shot down on 1 September; Andropov's circle feared a US first strike; NATO's Able Archer exercise was weeks away. Petrov was not a generic officer; he was an engineer-officer who understood the system's failure modes. His fifty-fifty self-rating after the call is part of the honesty of the case: he did not feel like a prophet. He felt uneasy, added base-rate reasoning (five is not a first strike; ground radar quiet), and chose. The A closer look refuses the cartoon "heart beats head." The synthesis is better: affective unease recruited the inferences that the protocol alone would have skipped.
From Willis to the triune demolition
Thomas Willis helped make the brain discussable in English; the Anne Greene case (1650) sits nearer his mythic anatomy than any gift of cadavers from Charles I. Darwin's questionnaires and baby notes built comparative emotion research and also licensed the vestigial reading. MacLean's triune model gave corporate trainers a diagram that will not die. The demolition is now consensus in neuroscience and still incomplete in popular culture. When an EI workshop draws three stacked brains, correct the anatomy without discarding the workshop's useful skills (naming feelings, practising reappraisal). The map can be wrong while some of the exercises remain good.
What Chapter 1 asks you to keep
Keep three habits. First, when evidence is ambiguous, name your mode before you name your conclusion. Second, treat gut unease as a hypothesis generator, not a verdict. Third, design (calendars, checklists, second confirmations) for the mode you will be in at the worst moment, not the mode you are in during the planning meeting. That is Petrov and Enterprise in one practice.
Ambiguous evidence as the emotion laboratory
Enterprise, shaken bees and Petrov share a structure: ambiguous evidence under stress. Alsbury's early unlock under vibration; bees refusing a mostly sweet mix after shaking; Petrov facing five missiles on a "highest reliability" system. Emotion's job in each case is to re-weight the ambiguity. Sometimes the re-weighting saves (melanoma mole; false-alarm call). Sometimes it kills (early feather unlock). Design and self-regulation both start by noticing: I am in a mode that changes what counts as signal.
- Petrov said later that his reasoning was partly logical: a real US first strike would involve hundreds of missiles, not five, and ground radar showed nothing. "Emotion versus logic" simplifies a mix of intuition and inference.
- The claim that Willis got cadavers "from King Charles I" is doubtful: Willis's anatomical work dates from the 1650s and 1660s, after Charles I's execution in 1649. His famous case, the hanged woman Anne Greene (1650), was a felon dissected under university privilege.
- The honeybee study (Bateson and colleagues, Current Biology, 2011) shows a judgement bias; whether bees feel anything remains disputed.
- The triune critique is now textbook consensus; MacLean's model survives mainly in popular books.
- Emotion and decisions. The integrative view (emotion informs rather than corrupts judgement) is mainstream. Large studies also confirm emotion's costs: incidental emotions (moods from unrelated events) still bias risk judgements.
- Insects. Bumblebee optimism-like and play-like findings (Solvi, Chittka and colleagues) strengthen the case for affective states in insects; the 2024 New York Declaration on Animal Consciousness called insect sentience a "realistic possibility."
- Triune brain. Reviews through 2025 continue to label MacLean's stack a neuromyth; it persists in popular EI training.
- SpaceShipTwo. Virgin Galactic flew commercial suborbital flights from 2023 and paused them in 2024 to develop its Delta-class ships.
You can test the chapter's claim on yourself. The next time you face an ambiguous decision (an unclear email from your boss, an odd noise at night), notice your emotional state first. Ask: if I were calm and rested, would I read this the same way? Petrov's lesson cuts both ways: a feeling that something is wrong deserves a hearing, but so does the possibility that anxiety is re-weighting the evidence. Keep a short log for a week of decisions made while anxious, angry or content, and compare how they turned out. Most people find their emotional reads were right more often than they expected, and wrong in predictable ways.
The old view pitted reason against passion; Darwin made emotions inherited but vestigial; the triune brain gave the error an anatomy. The new science shows emotion is a tool of thought: anxiety re-weights ambiguous data, love sustains parenting, and Petrov's gut feeling may have prevented a nuclear war.
2. The Purpose of Emotion
Mlodinow asks a hotel for a rushed beer and is told "Sorry, no." Two nights later he tries a different appeal and gets a better result. The contrast introduces the chapter's question: what does emotion add that reflexes cannot?
Reflexes, scripts and buttons
- Reflexes (the knee jerk) need a precise trigger and give one fixed response; they work even with the brain removed if the spinal cord is intact.
- Fixed action patterns or scripts are little programs for familiar situations, our "autopilot."
- Psychological buttons are intense learned reactions when someone ignores us, breaks rules or lies, dragging us back into unhealed issues.
Bacteria live entirely by scripts and do remarkably well; Antonio Damasio describes bacterial "cooperation," "snubbing" and aggression in competing flasks.
Student volunteers asked passersby in San Francisco and Santa Cruz for spare change. A routine request ("a quarter") triggered the refusal script. An odd request, "Can you spare thirty-seven cents?", broke the script, made people think, and raised giving markedly. Scripts govern more human behaviour than we realise.
Why emotion beats rules
In 1972 Allen Newell and Herbert Simon modelled thinking as a production rule system, a large set of if-then rules like a thermostat's ("If temperature < 70, turn on heat"). About a dozen rules capture children's subtraction. But avoiding bad food alone would need countless rules for smells, looks, hunger and alternatives. Rule systems fail when life is rich.
| Reflex | Emotion | |
|---|---|---|
| Trigger | Specific | General ("this smells funny") |
| Output | Fixed action | A mental state that biases processing |
| Timing | Immediate | Allows delay and strategy |
| Role of reason | None | Combines feeling with facts and goals |
Emotion lets reason enter the loop. Anger at an obstructive cable-company representative colours your thinking but need not become "go to hell, you idiot." In Frans de Waal's Chimpanzee Politics, a young male waits to mate out of the dominant males' sight. Caltech's David Anderson: at some point "organisms required more flexibility."
Do fruit flies feel?
Mlodinow compares his son Nicolai's account of a bar pickup ritual with fruit fly courtship: the male taps the female with a foreleg and "sings" by vibrating a wing. A mated female rejects him by extruding her ovipositor. Rejected males drink more alcohol, which flies find rewarding, much as people seek consolation.
Researchers had once catalogued 92 definitions of emotion. Anderson and Ralph Adolphs (2014) proposed testable properties instead:
- Valence: good or bad, approach or withdraw.
- Persistence: the state outlasts the trigger (after a snake, a rustling rodent makes you jump).
- Generalisability: many stimuli can cause it, and one stimulus can produce many responses.
- Scalability: intensity varies with circumstances.
- Automaticity: it arises without intention, but the response is not automatic.
These properties can be tested in animals that cannot report feelings.
Feeding flies saw an overhead shadow, a predator cue. They jumped away and returned after seconds. A second shadow made them stay away longer. The same trigger gave a different, stronger response: persistence and scalability, so a state, not a reflex. A fly brain has about 100,000 neurons (a millionth of ours) yet shows fear and aggression.
Persistence and scalability are what make emotion efficient: a snake on the trail rationally raises the odds of more snakes nearby, and a lingering state of alarm encodes exactly that update without any conscious calculation. A reflex would treat every rustle as a fresh, unrelated event. Individual flies, like people, also differ in how strongly and how long they react, an early hint of the emotional profiles of Chapter 8. Our emotions evolved for the savanna, not cities, so some reactions misfire today. Mlodinow's conclusion: emotion "is a gift." It speeds sense-making, feeds reasoning and connects us to others.
The ladder: reflex, script, button, emotion
Chapter 2's central diagram is a ladder of responses. A reflex (knee jerk) fires from a precise trigger with one fixed response and can work without the brain. A script is a little program for familiar situations: our autopilot, and the entire behavioural repertoire of bacteria. A button is an intense learned reaction (being ignored, being lied to) that drags present conflict into old issues. An emotion sits above all three: many triggers, flexible responses, and a place for reason to enter the loop. David Anderson and Ralph Adolphs (2014) list five properties that mark an emotion state: scalability (intensity can rise and fall), valence (good or bad), persistence (it outlasts the trigger), generalisation (it spreads to new situations), and global coordination (body, attention and action align). Those properties are what make emotion useful in a world that scripts cannot fully anticipate.
| Property | What it means | Why it matters |
|---|---|---|
| Scalability | Intensity varies | Mild worry differs from panic |
| Valence | Positive or negative | Directs approach or avoidance |
| Persistence | Outlasts the trigger | Sets a temporary mode of operation |
| Generalisation | Spreads to new cues | Learning transfers |
| Global coordination | Body, mind, action align | The whole organism shifts |
Why flexibility beats rules
Fixed rules fail when the environment changes faster than the rulebook. Emotion is evolution's answer: a temporary mode of operation that re-prioritises goals without rewriting every script. Fear narrows attention to threat and prepares escape or defence. Anger mobilises confrontation when a goal is blocked. Disgust steers away from contamination. Joy broadens the repertoire of thought and action (Fredrickson's broaden-and-build). The same neurotransmitter systems appear across phyla; what differs is the sophistication of the prediction machinery that deploys them.
Mlodinow's fruit-fly cases (with Anderson's lab) show that even insects display states with the five properties: persistent, scalable, valenced responses that generalise. The 2023 to 2024 FlyWire connectome mapped the adult Drosophila brain at synapse resolution; affective neuroscience can now ask which circuits implement which properties rather than whether flies have a "limbic system." The question "do fruit flies feel?" is less useful than "which emotion-state properties do they implement, and how?"
Buttons and the cost of old learning
Buttons are the dark side of learning. A colleague's tone of voice matches a parent's dismissal; a partner's lateness matches an old abandonment. The present is hijacked by a past script with emotion-level intensity. Recognition is the first regulation move: name the button, notice the body, ask whether today's facts match yesterday's story. Chapter 9's reappraisal tools start here.
REFLEX → SCRIPT → BUTTON → EMOTION. Fixed, programmed, triggered, flexible. Emotion earns its place by letting reason into a loop that pure rules cannot handle.
Treating every intense reaction as a "button" from childhood can become a way to dismiss legitimate present anger. Not every strong feeling is old baggage. Shadow: pathologising current injustice as your unresolved past, or romanticising every impulse as "authentic emotion," both miss the ladder.
Cross-domain (genuine): incident response
In SRE and trading-floor incident response, runbooks are scripts; automated failover is reflex; a blameless postmortem that still triggers shame is a button. Emotion states (calm focus versus panic) change which runbook steps get followed. Designing for two-failure tolerance (Enterprise lesson, Chapter 1) and for known buttons (this chapter) is the same engineering problem: assume the human in the loop will enter a mode, and design so that mode still succeeds.
Purpose expanded: flexibility as the evolutionary product
Chapter 2's argument is teleological in the careful biologist's sense: emotion exists because it solved a problem that reflexes and scripts could not. The problem is variability. A knee jerk can be hardwired because the tendon tap means one thing. A social life among primates, and later among language-using humans, means ten thousand things. Emotion states provide a temporary policy that generalises across novel cues without waiting for evolution to write a new reflex.
Scalability and persistence in ordinary time
Scalability explains why a mild worry and a panic attack can share a family resemblance and demand different responses. Persistence explains why a harsh email at 9:00 still colours the 11:00 meeting. Generalisation explains why a betrayal in one friendship makes the next friendship harder to enter. Global coordination explains why your gut, your gaze and your voice arrive together when you are furious. Adolphs and Anderson's list is not trivia; it is a checklist for asking whether a state in a fly, a dog or a human deserves the emotion label in the functional sense.
Buttons at work and at home
Buttons are learned emotion-intensity responses to social triggers. Being interrupted, being laughed at, being left on read, being told "you always..." can fire a reaction sized for a childhood dining table. The chapter's clinical observation is that otherwise healthy relationships burn cycles on buttons. A practical protocol: when intensity feels larger than the present fact, say so out loud ("that hit an old button; give me a minute"). The pause is not therapy theatre; it is inserting reason into the loop that Chapter 2 says emotion uniquely allows.
Flies, connectomes and the death of the limbic monopoly
If emotion required a mammalian limbic system, flies would be out. They are not out. Persistent avoidance, valence-dependent choices and state-dependent generalisation show up in Drosophila. FlyWire's synapse-level map (2023 to 2024) lets researchers ask circuit questions that used to be rhetorical. The limbic monopoly dies twice: once from mammalian network science (emotion is not in one place), once from comparative work (emotion-state properties appear without a limbic label). Popular books that still point at "your lizard brain" are using a metaphor that mis-teaches the anatomy.
Why rules still matter
The chapter does not say "burn the rulebook." Scripts and reflexes are efficient. Emotion earns its keep in the residual: the cases scripts miss. Aviation checklists, surgical timeouts and trading limits are scripts that exist because emotion modes are powerful. The right stack is reflex where possible, script where regular, emotion where novel, and meta-cognition (Chapters 8 and 9) where emotion itself needs steering.
Room service and the limits of scripts
Mlodinow opens Chapter 2 with a small hotel story: late-night room service promised in forty-five minutes for a simple beer; he did not want to wait. The anecdote lands the chapter's question: when does a fixed programme (the hotel's script for night orders) fail a flexible need, and what kind of response system would do better? Organisms face that problem constantly. Reflexes handle precise triggers. Scripts handle familiar situations. Buttons hijack the present with old intensity. Emotion is the flexible layer that lets goals shift when the script's forty-five minutes are the wrong answer.
Clinical psychologists see buttons daily: a partner's sigh matches a parent's contempt; a colleague's late reply matches an old exclusion. Cycles of conflict repeat because each side presses the other's button. The regulation move is not "never feel"; it is to notice the mismatch between today's facts and yesterday's programme. Fruit-fly work (Anderson) and the Adolphs-Anderson five properties give the scientific spine; the hotel beer and the button quarrel give the lived one.
- Anderson and Adolphs define emotion states functionally; the framework says nothing about whether flies have conscious feelings. Mlodinow sometimes blurs the two.
- The "thirty-seven cents" study (Santos, Leve and Pratkanis, 1994) is small; "pique technique" replications show modest and variable effects.
- Newell and Simon's production systems were not simply "failures." They led to cognitive architectures (Soar, ACT-R) still used today; the point is that pure rules scale badly.
- Fruit flies "came about forty million years ago" refers loosely to the Drosophila lineage; insects, and simpler affect-like states, are hundreds of millions of years older.
- The complete adult fruit fly connectome (FlyWire, about 140,000 neurons, published in Nature, October 2024) now lets researchers trace the circuits behind fly "fear" and aggression, including Anderson lab work on persistent internal states in the hypothalamus of mice and fly neurons.
- The 2024 New York Declaration on Animal Consciousness says there is a "realistic possibility" of conscious experience in insects, though scientists remain divided.
- The rule-system failure has a modern echo: hand-coded expert systems gave way to learned models; large language models are flexible precisely because they are not rule lists. Whether they have anything like emotion states is widely doubted.
Identify one script and one button in your own life. A script might be how you answer "How are you?" or reject a street fundraiser; a button might be how you react when someone interrupts you. Scripts save effort and are fine when the situation is routine. Buttons are emotional reactions out of proportion to the trigger, usually because the trigger echoes an old hurt. Mlodinow's advice is to learn other people's buttons so you avoid pressing them, and to work at deactivating your own: notice the trigger, name the old issue it recalls, and leave a pause before responding. That pause is exactly the delay emotion makes possible and reflexes do not.
Reflexes and scripts are cheap but rigid. Emotions add flexible, persistent, scalable states that bias how we process a situation and leave room for reason. Anderson and Adolphs's five properties make emotion testable even in flies.
3. The Mind-Body Connection
Simon at the fence
Mlodinow's father Simon helped lead the Jewish underground in the Częstochowa ghetto. One night he and three comrades dug under a barbed-wire fence to reach a bribed German driver. Simon's clothes snagged; by the time he was free, the truck was pulling away. As he stepped forward to run for it, something stopped him. It was not fear, he said; danger had become routine. Moments later an SS vehicle intercepted the truck and gunned down everyone aboard. Without that unexplained hesitation, Mlodinow would not exist.
His explanation: the unconscious processes information in the background (as when answers come in the shower). Simon's conscious mind weighed one set of facts, while his unconscious read subtle cues about the surroundings and his body, signalled through a built-in sense of well-being.
Core affect
Core affect is a continuous mental state reflecting your physical viability, "a kind of thermometer" of well-being built from body data, external events and thoughts. It is older and simpler than emotion and has only two dimensions:
- Valence: pleasant to unpleasant ("everything seems fine" or "something is wrong"), linked to the orbitofrontal cortex.
- Arousal: energised to sleepy, linked to the amygdala.
It is always present, like body temperature, but noticed only when you attend to it. Core affect is the body's input to every emotion, combined with situation, context and knowledge.
A lottery win lifts your mood for days but changes core affect less, because core affect tracks physical rather than financial well-being. Drugs, music, films and art all move it.
Life against entropy
Following Schrödinger (What Is Life?, 1944), life is a battle against entropy. Cells fend off disorder; in multicellular animals the nervous system maintains homeostasis. Core affect is a higher-level sentinel for threats to it.
In the 1980s Thomas Caraco offered captive dark-eyed juncos two trays of millet: a fixed amount or a variable amount with the same average. Warm birds chose the safe tray; cold birds, whose fixed ration would not keep them alive overnight, gambled. Humans are similar: someone who needs more income may take a riskier, better-paid job. The polite or rude way you answer a telemarketer on Saturday morning depends on breakfast.
The gut-brain axis
The mind-body link runs through nerves and through molecules such as serotonin and dopamine. Bacteria sensed their surroundings almost four billion years ago; multicellular life came about 600 million years ago, and early nerve nets mainly ran digestion (Damasio calls hydras "floating gastronomical systems"). Our enteric nervous system, the "second brain," runs the gut and talks to the brain through the gut-brain axis, in both directions.
- Transplanting gut microbes from timid mice into other mice made them more timid; microbes from adventurous mice made them bolder.
- Human MRI studies link brain connectivity to the dominant gut bacteria.
Why a head transplant would fail
In 1966 psychologist George Hohmann, paraplegic after a war injury, interviewed spinal-cord-injured veterans: the higher the injury (the less body feedback), the weaker their reported emotions. "Mind-body connection" got five Google hits for the 1960s and hundreds of thousands in the last decade. Surgeons Sergio Canavero and Xiaoping Ren proposed a $100 million head transplant in China. Bioethicist Paul Root Wolpe: an entirely new body would force a "massive reorientation" that could change the brain's very nature.
The brain as a prediction machine
Surprise reveals that the brain is always forecasting. These are immediate, personal predictions ("last time the bushes rustled, a bear came out"), and core affect supplies the body's side of them. His 98-year-old mother has lost reasoning but kept many of these fast predictions.
Kamal Abbasi, caught in a sting while buying explosive ingredients at 19, came before an Israeli parole board after five spotless years. His eleven-minute hearing was the last before lunch, and he was denied. In a classic study of 1,112 rulings by eight experienced judges, favourable decisions fell from about 65% at the start of each session to nearly zero before a food break, then jumped back. Similar effects: doctors prescribe more unnecessary antibiotics later in their shifts (21,000 visits, 200 clinicians), and analysts on earnings calls grow more negative as the day goes on. Low blood sugar made abusive men more aggressive; bad-tasting drinks made people judge others more harshly.
After a house fire left the family in a cramped flat for six months, Mlodinow learned to discount everyone's irritability. The lesson: our decisions, and other people's, are coloured by bodily states none of us notice.
Simon at the fence
Mlodinow's father, Simon, once hesitated at a camp fence when escape seemed possible. He could not later explain the hesitation in words. The chapter uses that moment to introduce core affect: the continuous, body-based sense of valence (pleasant to unpleasant) and arousal (calm to activated) that underwrites every feeling. Sometimes the body moves before a story arrives. Sometimes the story never arrives. Interoception (sensing the internal body) is the channel; prediction is the filter.
Core affect: the two axes
Core affect is not an emotion. It is the raw material. Plot valence on one axis and arousal on the other; every moment of waking life sits somewhere on that plane. High-arousal pleasant is excitement; high-arousal unpleasant is fear or rage; low-arousal pleasant is calm contentment; low-arousal unpleasant is gloom. Emotions are what happen when core affect is interpreted in a situation with the help of concepts and culture. That is why the same racing heart can be love on a date and terror in an alley (Schachter and Singer, Chapter 5).
| Low arousal | High arousal | |
|---|---|---|
| Pleasant | Calm, serene | Excited, elated |
| Unpleasant | Sad, bored | Afraid, angry |
Life against entropy
Schrödinger's What Is Life? (1944) frames organisms as local fighters against entropy: they maintain order by exporting disorder. The body is a continuous project of regulation (allostasis, not mere homeostasis). Feelings are the brain's readout of how that project is going. Hunger, fatigue, inflammation and gut signals are not distractions from thought; they are data. Hohmann's 1966 interviews with people who had spinal-cord injuries found that the higher the lesion (the less body signal reaching the brain), the more muted the reported emotional intensity. The body is not optional equipment for feeling.
The gut-brain axis
The gut's enteric nervous system and its microbiome send continuous signals via the vagus and immune routes. Mlodinow surveys evidence that gut state shifts anxiety and mood. The A closer look applies: much of the microbiome-emotion literature is early, rodent-heavy and overclaimed in popular press. The honest residue is that interoceptive channels include the gut, and that "rational" decisions made while nauseated or starving are decisions made under a shifted core affect.
Hungry judges and the A closer look
The famous Danziger et al. (2011) parole study reported that favourable rulings dropped toward zero before food breaks and jumped after. It became a textbook example of body state biasing justice. Later work (including Glöckner and scheduling analyses) showed the effect is overstated: case ordering, fatigue and judicial scheduling confound the simple "hungry = harsh" story. The broader claim survives: blood glucose, sleep and inflammation shift core affect, and core affect shifts judgement. The single dramatic graph does not.
Prediction machine
The brain does not passively receive body signals; it predicts them and updates on error. Predictive processing accounts of interoception (Barrett, Seth and others) treat feeling as the brain's best current guess about the body's condition, constrained by prior experience. That is why placebo, expectation and context can change pain and mood without changing the peripheral tissue. Chapter 5's constructionist story is the same architecture applied to named emotions.
- Predictive interoception remains a leading framework; debates continue over how much of "emotion" reduces to prediction error.
- The hungry-judges headline should be taught with Glöckner-style critiques and scheduling confounds, not as settled fact.
- Triune-brain language in corporate training is still common; treat as neuromyth when it appears.
BODY → PREDICT → AFFECT → FEELING. Signals rise; the brain guesses; valence and arousal result; context names the emotion. Simon's hesitation is the body voting before the story arrives.
"Listen to your body" becomes harmful when interoception is noisy (anxiety disorders, panic, somatisation) or when gut-mood claims are used to sell unvalidated microbiome fixes. Shadow: treating every bodily twinge as deep truth, or treating the body as irrelevant to reason, both fail the chapter.
Body first: expanding the mind-body chapter
Chapter 3 is the hinge of Part I. Without core affect, the later constructionist and regulation chapters have nothing to construct or regulate. The body is not a vehicle the mind rides; it is part of the mind's computational substrate.
Allostasis rather than a thermostat
Homeostasis suggests a fixed set point. Allostasis suggests anticipatory regulation: the brain predicts need and adjusts before the deficit arrives. Feeling hungry before the meeting that always runs long, feeling keyed up before a known conflict, feeling drained on the anniversary of a death: these are allostatic patterns, for better and worse. Predictive processing makes the same point in formal dress. Core affect is the summary statistic of how allostasis is going.
Hohmann, spinal injury and the intensity gradient
Hohmann's 1966 interviews are old and small by modern standards; their qualitative pattern still matters. People with higher spinal lesions, receiving fewer afferent signals from the body, reported more muted emotional intensity. The finding fits a family of results: interceptive accuracy varies across people; alexithymia (difficulty naming feelings) often co-travels with noisy or ignored body signals; panic disorder can be read as catastrophic interpretation of benign arousal. Chapter 5 will make interpretation central; Chapter 3 insists there is something to interpret.
Gut claims, calibrated
The gut-brain axis is real as anatomy (vagus, immune mediators, microbial metabolites). Popular claims that a specific probiotic will fix your marriage are not. If you notice mood shifting with illness, antibiotics, or hunger, you are observing Chapter 3, not discovering a personality flaw.
Judges, glucose and the replication ethic
Danziger's parole graph is catnip for textbooks. Later analyses showed case ordering and scheduling can produce similar curves without a simple hunger mechanism. Sleep deprivation, inflammatory illness and acute stress have stronger, repeated evidence as judgement shifters. Design workdays and court schedules for brains that have bodies.
Simon's pause as data
The fence story is not a miracle narrative. It is a single-case illustration of a decision completed before a verbal reason. Soldiers, surgeons, parents and traders all know such pauses. Training can reshape them (drills build better gut hypotheses); it cannot eliminate the body vote. The prediction-machine diagram in this chapter is the generalisation of Simon's night: signals, prediction, comparison, affect, feeling.
Simon's fence, second by second
Simon led underground work in the Czestochowa ghetto. On one night mission he and three others slipped under barbed wire toward a paid-off German soldier in a truck. A snag left Simon behind as the others moved. He had a moment: run and risk drawing fire on all of them, or stay and lose the mission. He began to step, then paused. He later told his son it was not fear; danger had become routine. Something in the body's vote stopped him. The truck left. He survived that night. The chapter refuses to romanticise the pause as destiny; it uses it as evidence that core affect and interoception can decide before a verbal reason arrives. Hohmann's spinal-injury interviews, Schrödinger's entropy framing and the gut-brain axis are the laboratory and theoretical parallels to that pause.
- Hungry judges. Weinshall-Margel and Shapard (PNAS, 2011) showed cases were not randomly ordered: the board finished one prison before a break, and unrepresented prisoners, who win less often, usually came last. Glöckner's simulations (2016) reproduced much of the pattern from sensible scheduling. The original authors say the effect survived controls for representation, but the size is almost certainly overstated, and the "depletion" explanation is weak: a large preregistered replication of ego depletion (Hagger and colleagues, 2016) found an effect near zero.
- Hohmann's small 1966 interview study did not replicate well; Chwalisz, Diener and Gallagher (1988) found spinal-cord-injured people reported emotions about as intense as others.
- Microbiome: the mouse transplant results are robust; causal claims about human personality from gut bacteria are not yet established.
- "Valence originates in the orbitofrontal cortex" is too neat; valence is represented across many regions.
- Interoception, the brain's sensing of the body's internal state, is now central to affective science. Predictive-processing accounts (Barrett, Seth, Friston) treat feeling as the brain's prediction of bodily state, a process called allostasis.
- A 2026 Trends in Cognitive Sciences review notes that interoceptive accuracy (for example, heartbeat counting) relates inconsistently to emotion and mental illness, and argues that what matters is how people interpret bodily signals ("interoceptive self-appraisal").
- Reviews in 2025 find the best evidence for cardiac interoception, with respiratory and gastric evidence still limited.
- Gut-brain research has moved to human trials of "psychobiotics" and microbiome-depression links; effects so far are small and inconsistent.
- Canavero's head transplant never took place.
Before any important decision, run a core affect check: Am I hungry? Tired? Cold? In pain? Just back from a stressful meeting? If yes, either fix the bodily state (eat, rest, walk) or delay the decision. Organisations can build this in: schedule hard cases early, add breaks to long review sessions, and rotate the order in which proposals are considered. When someone else is irritable, ask whether their body might be the cause before taking it personally, as Mlodinow did with his family after the house fire. Core affect is always present, but it influences us most when we fail to notice it.
Core affect (valence and arousal) is the body's constant report to the brain, rooted in homeostasis and the gut. It silently biases risk-taking and judgement, from juncos to parole boards, and it feeds the brain's continuous predictions. Noticing it is the first step to controlling its effects.
4. How Emotions Guide Thought
Dirac discovers feeling
Paul Dirac (born Bristol, 1902), who predicted antimatter, grew up under a tyrannical Swiss father who forced the children to speak only French at his table. He became famously silent and literal. Then he met Manci (Margit Wigner, sister of physicist Eugene Wigner), "his antiparticle": talkative, emotional, impulsive. Her letters were long and full of feeling; his were a few lines every few weeks: "I cannot write such nice letters to you... perhaps because my feelings are so weak." Slowly he changed: "You have made me human." They married in 1937. Dirac came to believe emotion helped his physics; in research, "A plus B could be C or D or E," and feeling guides which path to explore.
Emotion as a mode of operation
Jordan Cardella, 20, tried to win back his ex-girlfriend by having an acquaintance shoot him so she would pity him. The friend fired only once, into his arm; police saw through the story. Prosecutor: "the most phenomenally stupid case"; defence lawyer: "felony stupid."
"An emotion is a functional state of the mind that puts your brain in a particular mode of operation that adjusts your goals, directs your attention, and modifies the weights you assign to various factors as you do mental calculations" (Ralph Adolphs). Like an iPhone in low-power mode, the same hardware runs differently. Even "cold logic" runs inside an emotional mode.
Lost with his eight-year-old son in the Southern California desert, Mlodinow calmed himself, followed a hunch and walked out the right way: unconscious processing, delivered as a feeling.
| Emotion | Mode of processing |
|---|---|
| Fear | Heightened perception of threat; overestimates risks (after a grisly stabbing story, people rate all calamities, even natural disasters, as likelier) |
| Anger | Evolved for bargaining; makes risks seem smaller and others more to blame |
| Disgust | Expels and rejects; in social form, condemns |
| Hunger | Raises the value of acquiring, not just food |
| Sexual arousal | Lowers disgust and caution; raises willingness to take risks |
| Happiness, pride, interest, awe | Broaden attention; encourage exploration and connection |
| Sadness | Slows down and rethinks goals; more careful analysis |
In the early 1990s two exhausted hunters in their twenties, walking a logging trail near Bozeman, Montana at midnight, nervous about bears, saw a large object moving and grunting. One fired. The "bear" was a yellow tent; a woman inside died. The jury could not understand the mistake, but jurors were not tired and afraid. Perception passes through layers of interpretation, and fear tuned the interpretation.
Social emotions and morality
As our ancestors grew more social, emotions such as indignation, gratitude, admiration, jealousy, shame and guilt evolved to police loyalty, honesty and reciprocity. Jonathan Haidt's "The Emotional Dog and Its Rational Tail" (2001) argues that moral judgements come first from intuition and emotion, with reasoning added afterwards. Disgust, once a guard against bad food, became a guard against bad people: students judged moral violations more harshly in a room sprayed with fart spray or after a bitter drink (a filthy-room version, with greasy pizza boxes, failed).
Psychopaths, such as the 2017 Las Vegas and 2018 Thousand Oaks gunmen, lack empathy, guilt and shame. Their calculations are logical but unconstrained by social feeling. Psychopathy is linked to amygdala and prefrontal abnormalities and affects an estimated 0.02 to 3.3% of people.
Drives as emotions
Hunger, thirst, pain and sexual desire now count as emotion-like modes. After a disgusting Trainspotting toilet clip, Carnegie Mellon volunteers sold items more cheaply: disgust drives disposal, the reverse of hunger.
| Question | Not aroused | Aroused |
|---|---|---|
| Are women's shoes erotic? | 42 | 65 |
| Is a woman sexy when she's sweating? | 56 | 72 |
| Could you enjoy sex with someone you hated? | 53 | 77 |
| Can you imagine sex with a 60-year-old woman? | 7 | 23 |
Arousal also raised willingness to have unprotected sex, in women as well as men. Erotic films reduced women's disgust ratings.
Joy and sadness
In 1915 the Endurance was crushed by Antarctic ice. In 1916, starving and frostbitten, Shackleton's men set out from a cove on a near-suicidal crossing: "we were feeling happy. We even broke into song."
Barbara Fredrickson's broaden-and-build theory: negative emotions narrow attention for survival; positive emotions broaden it, encouraging play, exploration and bonds that build resources for later. Happy people are more creative and flexible; pride drives sharing and striving; interest drives learning; awe shrinks the self and turns us toward others. A 2010 review linked positive emotion to health through hormonal, immune and anti-inflammatory pathways; volunteers with more positive emotion were less likely to develop colds after rhinovirus nose drops. Modern "discord" (little play, exercise or nature) cuts positive emotion; daily gratitude and small pleasures help.
Sadness, "the architect of change," follows loss. It slows us down, signals need to others, and pushes us to rethink beliefs and goals; in a simulated foreign-exchange task, sad traders made more realistic, less overconfident judgements.
Dirac discovers feeling
Paul Dirac, emblem of icy rationality, once admitted that an equation's "beauty" guided him when evidence was thin. Mlodinow uses that confession to open the claim of Chapter 4: emotion is a mode of operation that changes what counts as a good reason. Joy, fear, anger and sadness do not merely colour thought; they change its parameters.
Emotion as mode of operation
Each major emotion sets a temporary policy for attention, memory and risk. Fear focuses on threat and favours safe options. Anger increases certainty and appetite for confrontation. Sadness slows search and can improve detail memory for what went wrong. Joy (Fredrickson's broaden-and-build) widens the thought-action repertoire and builds long-term resources. The Ariely and Loewenstein (2006) arousal study with Berkeley men showed large shifts in stated willingness for risky or ethically fraught acts when sexually aroused versus not: the same person, different mode, different answers.
Social emotions and morality
Guilt, shame, pride, gratitude and contempt are social emotions that regulate cooperation. They are not optional decorations on a rational ethics; they are part of how moral judgement is computed. Damage to emotion-related circuits (classic cases in affective neuroscience) can leave IQ intact while wrecking real-life choice: Damasio's work on patients who "know" but cannot feel the stakes is the clinical shadow of Dirac's beauty-sense.
Drives as emotions
Hunger, lust and pain are often parked outside "emotion" in folk psychology. Mlodinow, following the affective-science trend, treats them as cousin states: they have valence, arousal, persistence and global coordination. Separating "drive" from "emotion" is partly semantic. What matters for the book's thesis is that all of them re-weight thought.
Joy, sadness and the A closer look
Broaden-and-build is well replicated for mild positive affect and creativity measures; it is not a cure-all, and manic or forced positivity is not the same as joy. Sadness aids systematic processing in some paradigms and rumination in others. The chapter's honest takeaway: match the mode to the task. Brainstorm in a lighter mode; audit errors in a more careful, even sombre one.
MODE, NOT NOISE. Fear narrows, anger certifies, sadness details, joy broadens. Pick the mode for the job; do not pretend you are in none.
Modes are tools with side effects. Shadow: weaponising a mode (intimidation as "motivation") or refusing all negative modes as unenlightened.
Modes in the wild: expanding Chapter 4
Anger as certainty drug
Anger increases subjective certainty and narrows the option set toward confrontation or punishment. That is useful when a boundary is real and a cost must be imposed. It is disastrous in a probabilistic forecasting meeting. Organisations that reward visible anger as "passion" systematically over-select confrontational policies. The mode lens lets you ask a cleaner question than "is anger bad?": is confrontation the task this hour?
Fear, safetyism and missed risk
Fear's narrow attention catches tigers and misses slow risks (debt, climate, skill decay). A culture stuck in fear mode builds safetyism: it eliminates visible harms and incubates invisible ones. Chapter 4's prescription is not bravery posters; it is mode hygiene. Schedule threat review and opportunity review as separate meetings with different opening affect.
Sadness, audits and mourning
Sadness slows search and can improve memory for details of loss. That makes it a natural audit mode after failure, and a natural mourning mode after death (Epilogue). Forcing cheer in either context is not resilience; it is mode mismatch. Broaden-and-build belongs in ideation and relationship repair, not in the first hour after a production outage.
Joy and the resource bank
Fredrickson's broaden-and-build claims that mild positive affect widens thought-action repertoires and builds durable resources (skills, friendships, health behaviours). Evidence is solid for mild positive affect and creativity or openness measures; it is not a warrant for mandatory fun. Manufactured joy in a sick culture is another mode mismatch.
Social emotions as governance
Guilt, shame, gratitude, pride and contempt are how groups govern without police. Guilt repairs; shame excludes; gratitude binds reciprocal pairs; pride marks valued competence; contempt marks moral exile. Political and corporate cultures differ in which social emotion they weaponise. A guilt culture can correct errors; a shame culture can freeze learning. Chapter 8's inventories make this personal; Chapter 4 makes it structural.
Drives at the board table
Treating hunger, lust and pain as outside "real emotion" lets people pretend board decisions are affect-free. They are not. A hypoglycaemic steering committee is a Chapter 3 problem wearing Chapter 4 clothes. Name drives as modes when they are present. The Dirac lesson generalises: beauty, disgust, irritation and enthusiasm already sit inside "technical" judgement.
Beauty as data: Dirac and the affective parameter
When Dirac let "beauty" guide an equation, he was not abandoning physics; he was admitting an affective parameter into theory choice under uncertainty. Chapter 4 generalises that admission. Modes of operation change which arguments feel decisive. Traders, scientists, parents and judges all run under modes. The skill is not to eliminate the parameter; it is to know which mode you are in and whether it fits the task (brainstorm versus audit; confront versus repair).
- Disgust and morality. The "fart spray" and bitter-drink studies belong to a literature that has replicated poorly. A 2015 meta-analysis (Landy and Goodwin) found incidental disgust has a small effect on moral judgement that may vanish once publication bias is corrected. Mlodinow's "generally well replicated" is too generous.
- Arousal study. The table comes from Ariely and Loewenstein (2006), with about 35 men. Dan Ariely's later work faced a 2021 data-fabrication finding in an unrelated study; this one has not been challenged, but it is small.
- Broaden-and-build has support, but Fredrickson's famous "3:1 positivity ratio" (with Losada, 2005) was mathematically discredited and partly withdrawn in 2013.
- Psychopathy is not a separate DSM-5 diagnosis; it overlaps with antisocial personality disorder. The wide prevalence range reflects different definitions.
- Calling the mass shooters psychopaths is speculation.
- The appraisal-tendency framework (Lerner and Keltner) remains the main account of how specific emotions shift risk perception: fear raises risk estimates, anger lowers them. These are among the better-replicated findings.
- Awe research (Keltner, Awe, 2023) has grown fast, with "awe walks" improving daily mood in older adults.
- Post-replication-crisis work favours large, preregistered studies; many small "embodiment" effects (cleanliness, power poses, bitter tastes) are now treated with caution.
- The positive-emotion and health link holds in large cohorts, although causation remains hard to establish.
Name the mode you are in. Before a judgement, ask: am I afraid (and so overestimating risk), angry (underestimating it and blaming others), disgusted (rejecting), or happy (overconfident but creative)? Then use the emotion deliberately. Fear is useful for checking safety; sadness for honest post-mortems; happiness for brainstorming; anger for standing firm in a negotiation where you have been wronged. For important moral judgements, be wary of disgust: the strength of an "ugh" reaction is not evidence that something is wrong. And remember the Bozeman hunters: exhaustion plus fear is a dangerous combination for interpreting ambiguous evidence.
Every emotion is a mode of operation that re-sets goals, attention and weights. Fear sees bears in tents; anger bargains; disgust rejects; arousal lowers caution; joy broadens; sadness rethinks. Even physics, Dirac found, needs feeling to choose a path.
5. Where Feelings Come From
Mlodinow's father, in later life, paused every few steps because of angina: chest pain that warns a struggling heart to slow down. Nature is lenient about long-term health but strict in acute moments: you gag at faeces and recoil from a fierce animal.
Pain that surgery cured, without curing anything
Mid-century surgeons treated angina by tying off the internal mammary artery. Patients reported relief, yet autopsies showed no better blood flow. Two teams ran trials (unethical by today's standards) with sham operations: chest incisions but no ligation. About three-quarters of real-surgery patients felt better, and so did all five sham patients. One went from five nitroglycerin pills a day to ten in eight months. Pain did not track the degree of heart damage. The operation was abandoned.
Decades later, stents faced the same test: a trial with sham procedures (ORBITA, 2017) found stenting relieved stable angina little better than placebo. Cardiologists called it "very humbling." Placebos work through brain regions linked to emotion. The same body signal can produce very different feelings.
What determines an emotion
Psychologists Michael Boiger and Batja Mesquita describe Ann calling to say she will be late and Laura, who has spent days nursing her, snapping that it is irresponsible. Their reactions reflect the past few days and their history, not just the call. Being cut in line annoys you more when you are hungry.
Subjects were injected with "Suproxin," really adrenaline (racing heart, flushing, fast breathing) or saline. Some were told the side effects; others were not. They then waited with a confederate who acted either euphoric (paper aeroplanes, hula hoops) or angry (over an insulting questionnaire). Uninformed adrenaline subjects caught the confederate's mood; informed ones, who had an explanation for their arousal, did not; saline subjects felt little. Emotion = arousal + interpretation from context.
Perception is construction
The brain cannot process raw reality; it predicts and fills gaps, as in visual illusions and in speech, where we "hear" sounds that were masked. Wine tastes better when people think it is expensive. Emotion works the same way: circumstances, environment, mental state and core affect are integrated with shortcuts into a feeling.
The constructionists
Constructionists, led by Lisa Feldman Barrett, go further and question whether discrete emotion categories exist at all.
| Basic emotion theory (Darwin, Ekman) | Constructed emotion (Barrett) | |
|---|---|---|
| Categories | A few innate, universal emotions | Learned, culture-dependent concepts |
| Brain | Dedicated circuits ("fear centre") | Distributed networks used for many things |
| Body | Each emotion has a signature | No reliable fingerprints |
| Faces | Universally recognised expressions | Expressions vary with culture and context |
| Language | Labels name natural kinds | Labels help create the experience |
Evidence Mlodinow cites:
- Words like "fear" are catchall terms, as William James noted in 1894. Patients with amygdala damage felt nothing when snakes and scorpions crawled on them, yet panicked when breathing air rich in carbon dioxide: two "fears," two mechanisms.
- Categories also overlap (fear and anxiety).
- Cultures carve emotion differently: Turkish kızgınlık covers both sadness and anger; anger-like emotions are commoner in individualist Western cultures; the Chewong of the Malay Peninsula have about seven emotion words, English hundreds.
In an article co-written by Mlodinow, Barrett and Adolphs, Barrett argued science has found no objective criteria to detect a particular emotion in a person from body or brain alone.
Emotional intelligence
In 2018 a Thai driver, Nakharin Boonchai, hit an elephant near Khao Yai National Park; it turned and stomped his car, killing him. Animals clearly have strong emotion states. Humans add knowledge of our own feelings. His nearly 100-year-old mother cannot add nine and three, yet reads emotions perfectly.
Peter Salovey and John Mayer coined "emotional intelligence" in 1990: the ability to perceive, understand, use and manage emotions, in oneself and others.
In a 2008 experiment by Adam Galinsky and colleagues, MBA students negotiated the sale of a gas station where the buyer's limit was below the seller's minimum, so a deal required creativity. Those told to imagine what the other side was thinking or feeling reached creative deals more often than controls. In Galinsky's results, perspective-taking (thinking) beat empathy (feeling) for getting a deal.
We are wired to share: 30 to 40% of conversation and about 80% of social media posts concern our own experiences; in one study people gave up money to answer questions about themselves, and self-disclosure activates reward regions. Mlodinow contrasts two corporate bosses: an empathetic one who earned loyalty and a cold one who did not.
Pain that surgery cured, without curing anything
In 1959 to 1960, surgeons tested internal mammary artery ligation for angina. Real ligation helped many patients; sham operations (cut and sew without ligating) helped at least as often in the small sham group (five of five in one report). The operation was abandoned. The pain relief was real; the mechanism was expectation, meaning and context, not plumbing. Chapter 5 takes that shock as the door into constructed emotion: the same bodily ingredients, different story, different feeling.
What determines an emotion
Constructionists (Lisa Feldman Barrett and others) argue that an emotion instance is built from: core affect (valence and arousal), interoceptive prediction, situational context, and emotion concepts learned from culture and language. There is no fingerprint in the face or the amygdala that uniquely identifies "fear" across people and cultures. What looks universal is often the category label imposed by the experimenter's language.
Perception is construction
Visual illusions make the same point for sight: the brain builds a scene from incomplete data and priors. Emotion construction is the interoceptive analogue. Schachter and Singer's 1962 adrenaline study (same arousal, different confederate behaviour, different labelled emotion) is the classic psychology demo; modern critiques note weak effects and demand characteristics, but the broader constructionist programme does not rest on that one experiment.
Barrett versus basic emotion
The debate between basic emotion theory (Ekman tradition: universal, discrete emotions with evolved signatures) and constructed emotion (Barrett: categories built from affect and concepts) remains incommensurate as of 2025 to 2026. Barrett's replies to van Heijst and others defend the theory-ladenness of "universal" facial-expression findings. Basic-emotion researchers continue to publish cross-cultural recognition data. It reports Mlodinow's constructionist lean, flags the contested studies, and keeps the practical upshot both sides can share: context and concepts shape what you feel, and emotional granularity (having more precise words) helps regulation.
| Basic emotion (Ekman tradition) | Constructed (Barrett) | |
|---|---|---|
| Ontology | Discrete, evolved kinds | Categories built per instance |
| Face and body | Universal signatures | No unique fingerprints |
| Culture | Displays rules modulate expression | Concepts help constitute the emotion |
| Clinical upshot | Target the basic emotion | Build granularity and reappraisal |
| 2026 status | Still productive; universality claims contested | Influential; some find it hard to falsify |
Emotional intelligence, revisited
Salovey and Mayer's emotional intelligence (1990) and Goleman's popularisation assumed something like basic emotions you could perceive and manage. Constructionism revises EI: the skill is less "read the universal face" and more "build accurate concepts, track your affect, and choose regulation strategies." Granularity predicts better coping in several studies; forced labelling without acceptance can backfire (see Chapter 9).
- Barrett's research programme and critiques of facial-expression universality remain active; van Heijst-style challenges and Barrett's replies keep the debate open.
- Meta-analyses on emotion regulation (2025) find reappraisal and acceptance effective with culture as a moderator: strategies are not culture-free tools.
- Teach both frameworks; do not flatten the field into "science says construction wins."
AROUSAL + CONTEXT + CONCEPT = FEELING. Same heart rate, different story, different emotion. Surgery that cures by meaning is the parable.
Constructionism can be misread as "emotions are not real" or "you chose your anxiety." Constructed does not mean optional or blameworthy. Shadow: gaslighting someone's fear as a mere concept, or treating basic-emotion charts as destiny.
Construction under fire: expanding Chapter 5
What construction does and does not claim
Constructed emotion theory claims that instances of emotion are built, not that suffering is imaginary. The analogy is vision: seeing is constructive, and cliffs are still lethal. Critics worry that constructionism slides into "you create your own reality." Responsible constructionists (Barrett among them) deny that slide. Core affect is constrained by the body and the world; concepts shape the category the affect is placed in; culture supplies many of the concepts. You cannot concept your way out of hypoxia. You can concept your way from "I am dying" to "this is a panic sensation that peaks and falls," and that shift changes the instance.
Ekman's faces and the training problem
Basic emotion research trained observers on posed faces and then found high recognition. Constructionists argue that training teaches the categories the experiment later "discovers." Cross-cultural studies that avoid forced choice and posed faces often find messier results. The debate is methodological as much as ontological. For a reader, the actionable middle is clear: do not trust a six-face poster as a universal decoder ring; do invest in finer emotion words for your own states.
Schachter and Singer, handled with care
The 1962 adrenaline study is a teaching classic with replication and demand-characteristic worries. Use it as an intuition pump (same arousal, different label) and rest the theoretical weight on the broader literature: placebo analgesia, misattribution of arousal (bridge studies), and clinical work on panic reappraisal. Sham angina surgery remains the visceral case because the outcome was policy change, not only a p-value.
Emotional granularity as skill
People who differentiate "irritable," "betrayed," "exhausted" and "ashamed" cope better on average than people who only feel "bad." Granularity is learnable. It is also culturally inflected: some languages carve the affect plane differently. Chapter 9's labelling tool assumes some granularity; Chapter 5 explains why the words are not cosmetic.
EI after constructionism
Goleman's EI sold partly because it promised a soft skill with hard career returns. Meta-analyses on EI and job performance are mixed and construct-confused (mixed models vs ability models). A constructionist revision of EI emphasises: accurate affect tracking, concept richness, context sensitivity, and regulation choice. It de-emphasises: reading universal microexpressions as a party trick. That revision is more modest and more useful.
Sham surgery and the ethics of meaning
The angina ligation trials did not only overturn a procedure; they showed that meaning can be analgesic. That fact is clinically precious and ethically sharp. Placebo-controlled trials exist because expectation heals and also deceives. Constructionism inherits both edges: if concepts help constitute emotion, then teaching finer concepts can help, and manipulative framing can harm. Emotional intelligence revised along constructionist lines is a literacy of concepts and contexts, not a checklist of universal faces.
- Schachter-Singer is famous but shaky: attempts to replicate it (Marshall and Zimbardo, 1979; Maslach, 1979) found unexplained arousal tends to feel negative, whatever the confederate does.
- The sham-surgery trials were tiny (the "all five" sham patients).
- The amygdala-and-CO₂ patients include the famous patient S.M. (Feinstein and colleagues, 2013), whose panic suggests interoceptive fear bypasses the amygdala.
- Mlodinow co-wrote with Barrett, and the chapter leans toward her view. Basic-emotion researchers (Ekman, Keltner, Cowen) argue that cross-cultural studies show above-chance recognition of many expressions.
- Linguistic claims (Turkish kızgınlık, seven Chewong words) come from older ethnographies and are simplified.
- The debate is softening. Reviews in Perspectives on Psychological Science (van Heijst, Kret and Ploeger, 2025) argue the two camps answer different questions: basic emotion theory addresses evolution and function, constructionism addresses mechanism and feeling, and they may be complementary.
- Brain data mostly favour distributed networks over emotion-specific "centres": fMRI work (2022 onward) finds shared neural bases across anger, sadness and happiness, with individual-specific patterns. Machine-learning studies can classify emotion categories above chance, but the patterns do not generalise well across people and situations.
- Faces: a major 2019 review (Barrett and colleagues) found facial movements are unreliable signals of emotion. Commercial emotion-recognition AI is now restricted: the EU AI Act bans it in workplaces and schools (in force from February 2025).
- Placebo and stents: ORBITA-2 (2023) found that stenting did reduce angina compared with a placebo procedure when patients stopped anti-anginal drugs, refining the 2017 result.
- Emotional intelligence: ability-based EI (MSCEIT) predicts some job and social outcomes modestly; popular claims that EI matters more than IQ are not supported.
Because feelings are constructed, you can sometimes re-label them. A racing heart before a talk can be read as fear or as excitement. Studies by Alison Wood Brooks (2014) found that people who told themselves "I am excited" before singing or public speaking performed better than those who tried to calm down. Building a richer vocabulary for emotions (emotional granularity, in Barrett's term) also helps: people who can distinguish frustration from disappointment or anxiety from dread tend to regulate better and drink less under stress. In negotiations, follow Galinsky: think about what the other side is thinking, not only what they feel.
Feelings are constructed: the brain combines body signals, situation, history and culture into an emotion, just as it constructs what we see. The same angina, adrenaline or racing heart can feel like nothing, relief, euphoria or rage. Constructionists question fixed emotion categories; emotional intelligence is the skill of reading these constructions in ourselves and others.
6. Motivation: Wanting Versus Liking
In Derby, England, Clara Bates was evicted twice in a year because her toddler Farrah was eating the carpets and walls. Farrah had pica (Latin for magpie), first described in 1563: people crave detergent, sponges, chalk, soil. Why do we eat pasta and not pillows? Motivation is willingness to spend effort on a goal. It runs through the brain's reward system, a flexible way to weigh many factors and choose what to pursue. Understanding came in two leaps, decades apart.
Leap one: the "pleasure centre"
A psychiatrist, neurologist and psychoanalyst, Heath worked on a Columbia project to "improve" the lobotomy (1948). Believing in biological psychiatry, he thought schizophrenia stemmed from a faulty pleasure system. A chance chat on the beach at Atlantic City with Tulane's medical dean got him a job; from 1949 he ran a 150-bed unit at New Orleans' Charity Hospital and implanted deep electrodes in patients. Two of his first ten patients died. He dropped the schizophrenia theory in 1955 because it failed, not because of the human toll. In a 1972 paper he stimulated the pleasure region of a gay man ("B-19") while showing him heterosexual films and later hired a prostitute for him, claiming a "conversion." Mlodinow: "smart and creative, a visionary... but sloppy and reckless."
New postdoc James Olds, taught by Peter Milner at McGill, mis-aimed an electrode in a rat's brain. The rat kept returning to the corner where it was stimulated. Given a lever, rats pressed it thousands of times an hour, ignoring food and water, until exhausted. Olds and Milner concluded that pleasure is the main motivator, overturning drive theory (which said we act chiefly to reduce discomfort such as hunger).
Reward systems are ancient. The 302-neuron worm C. elegans uses dopamine; planaria (about 560 million years ago) had rudiments. Vertebrates share an all-purpose reward network, which is why an orange satisfies and a Cessna 150 does not. By the mid-1980s textbooks taught that dopamine is the brain's pleasure chemical. But addicts kept taking drugs they said they no longer enjoyed.
Leap two: Berridge
Kent Berridge (Michigan), once a student of Roy Wise's dopamine-pleasure theory, destroyed rats' dopamine neurons and found they stopped seeking food (they would starve) but still showed "liking" faces (tongue protrusions) when sugar was put in their mouths. Stimulating wanting circuits made rats eagerly eat food they did not like. Two separate systems:
- Wanting (incentive salience): dopamine, centred on the nucleus accumbens, compact, ancient, found in all animals.
- Liking (pleasure): opioids and endocannabinoids in tiny scattered "hedonic hotspots," evolved later.
Dopamine makes us want things; it does not make us enjoy them.
Barriers make us want more but like less. In a 2013 Hong Kong speed-dating study, men who met a woman who was hard to get (in fact a confederate told to be unresponsive) liked her less but wanted to see her again more.
Berridge also found the accumbens works like a keyboard of desire and dread: stressful settings (bright lights, loud noise) expand the dread zone; calm, familiar settings expand desire. A friend in a noisy office felt constant low-level anxiety without knowing why. For about 15 years nobody, including Wise, accepted Berridge's theory, and he worked without funding.
Buchenwald and the overeating instinct
Mlodinow's father entered Buchenwald in late 1943 at 165 pounds and weighed half that by spring 1945. Hiding with a friend, Moshe, as the SS evacuated prisoners, he emerged when US troops liberated the camp on 11 April 1945 at 3:15 p.m. The soldiers offered chocolate, salami and cigarettes. His father ate cautiously; Moshe gorged and died. In extremes, the reward system's settings fail.
Supernormal stimuli
Niko Tinbergen found male sticklebacks attacked crude red-bellied dummies more fiercely than real rivals; geese tried to roll a volleyball into the nest instead of their own egg. These supernormal stimuli beat anything natural.
- Drugs: chewed coca leaves are mild; refined cocaine and crack are highly addictive. Opium became refined, fast-acting opioids.
- Food: we evolved to like scarce sugar and fat. "Food optimisers" ("I've optimized pizzas... salad dressings and pickles") engineer hyperpalatable products. Rats with 24-hour access to Sara Lee cheesecake, frosting and pound cake overate and kept eating despite electric shocks, like drug-addicted rats.
- Obesity has been estimated to kill about 300,000 Americans a year. Addiction is now defined by behaviour (wanting overpowering liking), not only chemistry; susceptibility links to dopamine receptor genes.
Leap one: the hunt for the pleasure centre
In the late 1940s Robert Heath at Columbia and then Charity Hospital in New Orleans implanted deep electrodes in psychiatric patients, chasing a theory that schizophrenia was a pleasure deficit. In 1953 to 1954, James Olds and Peter Milner at McGill found that rats would press a lever thousands of times an hour to stimulate a midbrain site later tied to the mesolimbic dopamine pathway. The press called it a "pleasure centre." Heath's later work, including the infamous B-19 case (1972), seemed to confirm that humans would seek the same stimulation. The first leap of Chapter 6 is this discovery story: a button for bliss appeared to exist inside the brain.
Leap two: Berridge separates wanting from liking
Kent Berridge (University of Michigan) broke the leap. Dopamine, he showed, tracks wanting (incentive salience: the pull toward a cue or reward), not liking (the hedonic pleasure of consuming it). Liking depends more on opioid and related hotspots in the nucleus accumbens and ventral pallidum. Rats with dopamine depleted still make liking faces to sugar but will not work for it. Amphetamine and many addictive drugs inflate wanting far out of proportion to liking. That is why addicts can chase a drug they no longer enjoy. The folk phrase "I want it so I must like it" is false in the brain's bookkeeping.
| Wanting | Liking | |
|---|---|---|
| Rough substrate | Dopamine incentive salience | Opioid (and related) hedonic hotspots |
| Behaviour | Approach, craving, "must have" | Pleasure faces, enjoyment |
| Can dissociate? | Yes: high want, low like (addiction) | Yes: like without chasing |
| Hijacked by | Cocaine, amphetamine, cues | Some opioids; natural sweets |
| Book's lesson | Craving is not evidence of joy | Enjoyment can exist without obsession |
Berridge and Robinson's incentive-sensitisation theory marked roughly thirty years of evidence in a 2025 review: wanting can sensitise with repeated drug use while liking stagnates or falls. Opioid-crisis work through 2026 continues to show cue-triggered craving as a wanting phenomenon. Popular "dopamine detox" advice often confuses wanting with liking and oversimplifies tonic dopamine. Keep the dissociation; drop the detox folklore.
Buchenwald and the overeating instinct
Mlodinow links his father's camp experience to later patterns around food: starvation can recalibrate wanting for calories long after liking and need have normalised. The mechanism is not moral failure; it is a wanting system tuned by extreme scarcity. Similar patterns appear in famine survivors and in some dieting cycles: the cue of food becomes louder than the pleasure of eating.
Supernormal stimuli
Niko Tinbergen showed that birds will prefer an oversized, garish dummy egg to their own. Humans build supernormal versions of ancestral rewards: refined cocaine from chewed coca; fast opioids from raw opium; engineered hyperpalatable foods from scarce sugar and fat; infinite-scroll feeds from social grooming. The wanting system, evolved for sparse rewards, meets stimuli it cannot easily saturate. Liking may plateau; wanting does not. Design ethics for product teams and for personal habits starts here: every supernormal cue is a bet against your own incentive system.
Practical separation
Ask of any craving: do I want this, or do I like this? If wanting is high and remembered liking is low, you are in Berridge territory. Delay, change the cue environment, or substitute a liked-but-less-sensitising reward. If liking is high and wanting is calm, enjoy without the addiction narrative. Traders, product managers and anyone with a phone live in a supernormal ecology; the dissociation is daily hygiene, not only clinical.
WANT ≠ LIKE. Dopamine pulls; opioids enjoy. Addiction is wanting without liking. Supernormal cues are designed to widen the gap.
"Dopamine bad" becomes a moral panic that pathologises ordinary motivation. Wanting is how you finish a marathon and how you open a textbook. Shadow: shaming all desire as addiction, or treating every liked activity as safe because "I enjoy it" while cues silently sensitise wanting.
Cross-domain (genuine): product design and banking
Infinite scroll, variable rewards and streak counters are supernormal cue machines. The ethical test from this chapter: does the product inflate wanting beyond liking, and who captures the surplus?
Wanting systems in a supernormal world: expanding Chapter 6
Heath, ethics and the first leap
Robert Heath's deep-electrode programme sits in a grim ethical history: vulnerable patients, grand theories of schizophrenia, sexual orientation interventions that read as abuse by later standards. The scientific residue the book needs is narrower: humans will work for stimulation of mesolimbic sites, and the press called that pleasure. Berridge's later correction does not erase the ethical stain; it does correct the mechanism story. Readers should hold both: the history is cautionary, the wanting/liking split is the payload.
Incentive sensitisation in daily cues
Robinson and Berridge's incentive-sensitisation theory says repeated drug use can make cues hyper-salient even as pleasure fades. The daily analogue is milder and still real: the phone buzz, the fridge light, the "one more episode" autoplay. Sensitisation is not inevitable; cue management (leave the phone in another room; shop from a list; decide the number of episodes before pressing play) is wanting hygiene. Liking can be savoured when wanting is not inflamed.
Food, famine and family history
Mlodinow's link from Buchenwald scarcity to later food patterns is personal and biologically plausible. Famine and extreme diet cycles retune incentive salience toward calories. Shame-based diet culture then piles Chapter 8 shame on Chapter 6 wanting. The kinder, more accurate frame: your wanting system learned a world of scarcity; supernormal food cues exploit that learning; redesign cues before you moralise.
Product ethics
Variable rewards, streaks, infinite scroll and loot boxes are applied incentive salience. The chapter gives product managers a test: are we inflating wanting past liking, and do users regret the time after liking collapses? The cross-domain take is genuine because the mechanism is the same.
Pleasure without chase
The quiet upside of the dissociation: you can like without wanting obsessively. A walk, a meal, a piece of music can be scheduled as liking practice: reduce cues that inflame chase, keep the hedonic event. Addiction treatment already uses versions of this; ordinary life under-uses it.
Pica, Farrah and the question of motivation
Chapter 6's cold open is Farrah, a toddler in Derby whose craving to eat odd non-foods (pica, named from the magpie) got her mother evicted twice. Pica is an extreme of motivation aimed at the wrong target: the willingness to put effort into a goal that is not nutritive. The chapter's question follows: why pasta and not pillows? Why effort toward any goal? Biological drives, learned cues and incentive salience all answer parts of it. Berridge's wanting/liking split is the mechanism that makes Farrah's case and Heath's electrodes and Olds-Milner's levers speak one language: motivation is not identical with pleasure, and pleasure is not identical with the chase.
- Olds and Milner's paper title names the septal area; the exact sites that support self-stimulation, including the medial forebrain bundle, are wider than "the accumbens."
- Moshe's death after liberation was most likely refeeding syndrome, a metabolic collapse, not a failure of motivation.
- The "300,000 deaths a year" figure (from about 1999) is contested; later estimates range widely depending on method.
- "Food addiction" remains debated as a clinical diagnosis, though the Yale Food Addiction Scale finds it in a minority of people.
- Heath's B-19 study is widely regarded as unethical "conversion therapy" research; the chapter is right to condemn it.
- GLP-1 drugs (semaglutide, tirzepatide) quiet "food noise," cutting the wanting for food, and early trials suggest they may also reduce alcohol and other cravings. Large trials of GLP-1 drugs for alcohol use disorder are under way.
- Dopamine is now usually described as signalling reward prediction error and incentive salience; Berridge's wanting-liking distinction is mainstream, though some researchers argue hedonic hotspots are less localised than first thought.
- Ultra-processed foods: a 2019 NIH inpatient trial (Hall) found people ate about 500 kcal/day more on an ultra-processed diet; 2024 to 2025 umbrella reviews link high intake to many health harms.
- Adult obesity in the US was about 40% in the latest national survey (2021 to 2023).
Check whether a craving is wanting or liking. Before reaching for the phone, snack or purchase, ask: will I actually enjoy this, or do I just feel pulled toward it? Rate the pleasure afterwards; many people discover that habits they crave no longer deliver much liking. Change the environment rather than relying on willpower: remove cues (notifications, visible snacks), add friction, and seek calm, familiar settings, since stress expands the brain's "dread" zone and pushes us toward quick fixes. Be especially wary of products engineered as supernormal stimuli, from ultra-processed food to infinite-scroll apps.
Heath and Olds found a "pleasure centre"; Berridge showed it is really two systems: dopamine-driven wanting and opioid liking. Addiction, hyperpalatable food and hard-to-get partners pull them apart, so we can crave what we do not enjoy. Supernormal stimuli exploit an ancient system built for scarcity.
7. Determination
Buster Douglas
James "Buster" Douglas, a 42-to-1 underdog, fought undefeated champion Mike Tyson for the heavyweight title. Promoters were already planning Tyson against Evander Holyfield. Three weeks earlier Douglas's mother, Lula Pearl, who had been telling everyone her son would beat the champ, died of a stroke at 47. With five seconds left in the eighth round, a Tyson uppercut dropped Douglas; he rose at nine, "knocked out on his feet." In the tenth he knocked Tyson out. Asked how, he wept: "My mother." Once Douglas showed Tyson could be outlasted, others beat him too.
Muhammad Ali said he did "nine or ten" push-ups, because he started counting only when it began to hurt. Motivation is why we want something; determination is firmness of purpose in pursuing it.
Where determination comes from
In 1957 a 14-year-old Chilean boy, Armando, had brain surgery after repeated headaches. Afterwards he was completely apathetic: no voluntary movement, no gaze, indifferent even to awkward positions. As the swelling subsided after about a month, his drive returned suddenly and he called out to his parents.
Determination is not in one spot but in networks:
- The emotional salience network, with nodes in the insula, anterior cingulate cortex and amygdala, monitors body and world and flags what matters.
- The executive control network, centred on the prefrontal and parietal cortex, then keeps you focused on the goal and ignores distractions.
Salience says "this matters"; executive control says "keep going."
In 2013 Stanford neurologists led by Josef Parvizi, mapping an epilepsy patient's seizures, stimulated a tiny node in the anterior midcingulate cortex. The man felt "determination," like driving into a storm with a flat tyre, knowing you will make it. Moving the electrode a few millimetres made the effect vanish. Beta-blockers, which quiet salience, make people sluggish.
In a tube test, two mice meet in a narrow tube and the more dominant one pushes the other out. Researchers used optogenetics to switch on neurons in the dorsomedial prefrontal cortex of losing mice. The losers suddenly pushed harder and beat previous winners, and some kept winning after the light was off. Determination can be switched on and can build on success.
Parvizi suggests innate differences in executive control might be spotted in childhood. Two ways to build it:
- Aerobic exercise: as little as 15 minutes a day improves executive function.
- Mindfulness meditation: two weeks of training cut smoking by about 60% in one study.
The apathy of computers
Sophia, Hanson Robotics' 2015 humanoid robot (face modelled on Audrey Hepburn), charms with canned lines but decides nothing; it executes instructions from activation to shutdown. Animals constantly weigh costs and benefits and initiate action. That self-starting quality is what emotion provides.
Eighteen statements ("I am interested in things," "I have initiative," "I have friends") are rated 1 (very true) to 4 (not at all true). High totals mean apathy, low totals mean determination. The minimum is 18; healthy young adults average 24; people in their sixties average 28; about half score within four points of the average.
Declining will
Frontotemporal dementia erodes the salience network, causing severe apathy. Mlodinow's mother, once "opinionated, histrionic," kept a living room so sacred the children could enter only on Passover and the High Holidays. In assisted living she began eating bacon. Now, asked what she wants, she just smiles.
Sleep loss blunts what matters. REM sleep activates salience nodes, and one night without sleep amplified amygdala reactions to negative pictures by about 60%.
Buster Douglas and the night determination won
On 11 February 1990 in Tokyo, Buster Douglas, a 42-to-1 underdog, fought Mike Tyson. Douglas's mother, Lula Pearl, had died at 47 weeks earlier. In round 8 he was knocked down and rose at nine. In round 10 he knocked Tyson out. Asked what carried him, he said: "My mother." Chapter 7 opens with that night because determination is not a vague virtue; it is a state that can be triggered, sustained and, in the lab, even switched.
Where determination comes from
Josef Parvizi and colleagues stimulated mid-cingulate and related sites in human patients and elicited reports of will to push through challenge. In mice, a 2017 optogenetic study on the tube test (two mice push; the dominant one wins) found that activating dorsomedial prefrontal neurons in losers made them push harder and sometimes keep winning after the light was off. Determination, in this framing, binds a salience network (what matters now) to executive circuits (effort and control). It is not a single "willpower organ," and it is not merely dopamine wanting: you can want something and still quit, or persist at something you do not currently like.
| Layer | Role | Failure mode |
|---|---|---|
| Salience | Tags the goal as what matters | Distraction; everything feels urgent |
| Executive control | Allocates effort, inhibits quit signals | Fatigue; depletion narratives |
| Meaning / identity | "For my mother"; values | Empty goals collapse under pain |
| Body state | Glucose, sleep, pain | Hungry-judge territory (Ch 3) |
The apathy of computers
Mlodinow notes that machines optimise objectives without caring. Determination is the affective force that keeps a human on a hard objective when gradients are flat and rewards are delayed. AI systems can mimic persistence with search budgets and reinforcement learning, but they do not feel the cost of effort. The chapter's cross-domain edge for AI is honest only this far: objective functions are not determination; they lack the salience-plus-meaning stack that makes Buster Douglas rise at nine.
Declining will and the depletion debate
Popular psychology sold ego depletion (willpower as a muscle that tires) hard in the 2000s. Replication crises cut the effect size deeply; meta-analyses disagree on whether a small effect remains. Mlodinow's safer claim does not need the strong depletion story: determination fluctuates with sleep, glucose, meaning and recent wins; you can design days to protect it (hard tasks early, cues that renew salience, identity reasons that outlast mood). Declining will with age or depression is clinically real even if the undergraduate glucose-muscle model is not.
- Human stimulation and rodent optogenetic work continue to implicate mid-cingulate / dorsomedial prefrontal effort circuits; effect sizes and translation to healthy volunteers stay cautious.
- Ego-depletion as originally sold remains unreplicated at advertised size; teach fluctuation and design, not "willpower sugar."
- AI "agent" persistence is still search budget and RL reward, not affective determination. Do not anthropomorphise the loss curve.
SALIENCE + EXECUTIVE + MEANING. What matters, the effort to pursue it, and why it is yours. Douglas rose for his mother; the tube-test mouse pushed when the circuit fired.
Determination culture becomes grind culture: celebrate rising at nine while ignoring broken hands and broken sleep. Shadow: using "mindset" to deny structural barriers, or stimulating effort pharmacologically without the meaning layer that makes persistence sane.
Try-it deepened
Pick one hard task this week. Before starting, write one sentence of salience (why it matters today), one of meaning (who or what it is for), and one body check (sleep, food, pain). Mid-task, when quit signals rise, re-read the three lines before negotiating with yourself. That is Chapter 7 as a practice, not a slogan.
Determination expanded: circuits, stories, limits
Parvizi and the felt will
Patients who can report the effect of stimulation give determination research something rodent work cannot: the feeling of will. Mid-cingulate and related effort regions, when engaged, produce reports of rising to a challenge. That does not mean a surgeon should implant electrodes for grit. It means determination is not only a cultural sermon; it has a neural handle.
Tube tests and lasting winners
The 2017 optogenetic tube-test result matters because losers became winners and sometimes stayed winners after the light was off. Success can re-tune the circuit. That is the laboratory cousin of confidence born from a hard-earned win. It also warns against fixed labels ("he's just not a closer"): recent effort history shapes the next push.
Ego depletion's afterlife
Ryan and Deci's self-determination work, Dweck's mindset research and the depletion debate all orbit the same folk problem: why does effort fail? The post-replication answer is plural: sleep, glucose extremes, boredom, unclear goals, missing meaning, cue-driven distraction, depression. A single muscle metaphor was always too neat. Chapter 7's usable residue: protect the stack (salience, executive resources, meaning, body), and stop moralising every quit signal as character failure.
Machines that do not care
An optimiser with a long horizon looks determined. It is not. The difference shows in tradeoffs humans make when meaning conflicts with reward: raising a child with a disability, keeping a vigil, refusing a profitable harm. AI persistence without affective cost will keep beating humans at Go and keep missing why Douglas rose at nine. The honest AI take stops there; science fiction begins after it.
Training determination without grind culture
Practical training: pick goals with identity reasons; break them into salience tags for today; defend sleep; rehearse quit-signal scripts ("re-read the meaning sentence, then renegotiate"); celebrate process wins that re-tune the circuit. Refuse workplaces that treat determination as infinite unpaid overtime. The The dark side of this chapter is how grind culture steals a scientific idea.
Rising at nine: identity as fuel
Douglas's "My mother" is identity fuel. Tube-test mice do not have eulogies; they have circuits. Humans bind circuits to stories. Determination interventions that ignore story (only cold showers, only stimulants) miss the stack. Interventions that only tell stories without protecting sleep and salience also fail. The chapter's practical synthesis is three-layered practice: body ready, goal tagged as salient, meaning available when pain peaks.
- The 2017 mouse study (Zhou and colleagues, Science) came from Hailan Hu's lab in China, not "the Stanford group"; the book's transition blurs the two studies.
- Parvizi's stimulation involved two patients; the finding is striking but tiny.
- The mindfulness-smoking study (Tang and colleagues, 2013) had about 60 participants; larger mindfulness trials show smaller, inconsistent effects.
- The inventory is the Apathy Evaluation Scale (Marin, 1991), a clinical tool, not a measure of "grit."
- "Determination" overlaps with Angela Duckworth's grit, whose ability to predict success beyond conscientiousness has been questioned (Credé and colleagues, 2017).
- Sophia has been overtaken by large language models and AI agents that hold fluent conversations and carry out multi-step tasks. They still lack anything like the salience-driven, self-initiated goals Mlodinow describes: they act when prompted or when a programmer sets a goal.
- Apathy is now recognised as a distinct, treatable target in Parkinson's disease and dementia; trials of methylphenidate for apathy in Alzheimer's (ADMET 2) showed modest benefit.
- Exercise remains one of the best-supported ways to improve executive function and mood; a 2024 BMJ network meta-analysis ranked walking, jogging, yoga and strength training as effective for depression.
- Sleep research confirms that sleep loss amplifies negative emotion and reduces positive affect (a 2024 meta-analysis of over 150 studies).
To build determination, work on the networks rather than on willpower slogans. Move: 15 to 30 minutes of aerobic exercise most days improves executive function. Train attention: brief daily mindfulness practice strengthens focus and emotion regulation. Sleep: protect seven to nine hours, because sleep loss makes goals feel less important and threats more intense. Start small and win: the 2017 mice that won once kept winning, and early successes build momentum. Finally, connect the goal to something that matters emotionally, as Buster Douglas connected his fight to his mother; salience fuels persistence.
Determination is an emotion of persistence, produced when the salience network tags a goal as important and the executive network sustains effort. Damage brings apathy; stimulation brings an abstract will to persevere. Exercise, mindfulness and sleep strengthen it; age and exhaustion erode it. Machines that only follow scripts lack it.
8. Your Emotional Profile
Jim and his shame
Los Angeles psychiatrist Gregory Cohen: "Every person is unique... also emotionally." His patient Jim, on his third marriage, was baffled when his wife left: "We had a fine marriage." He admitted affairs ("all her fault"), called her an alcoholic, and blamed her for everything. Mlodinow rolled his eyes. Cohen: "He's not lying; he's just wrong." Jim's conscious mind believes he was a wonderful husband; underneath, a mother who constantly shamed him (shrieking and throwing him into his crib when he bit while nursing) left him hypersensitive. Shame is his "dominant emotional state," and denial protects him from it.
Your emotional profile is your personal pattern of emotional tendencies. For each emotion it has four parameters:
- Threshold: how much it takes to trigger it.
- Latency to peak: how fast it builds.
- Amplitude: how intense it gets.
- Recovery time: how long it takes to fade.
Together they form "a kind of emotional fingerprint." There is no right or wrong profile, only tendencies to know and, if you wish, to adjust.
Nature and nurture
Mlodinow's mother insisted she knew how each son would turn out: the elder shy and anxious, the younger chatty, Leonard "curious... in the dangerous manner." Infants of two to three months already differ in emotional reactivity. But experience matters too: Jim was "a prisoner of his childhood."
In the 1990s Michael Meaney at McGill noticed that rat pups of mothers who licked and groomed a lot grew up mellow (exploring new places, releasing little stress hormone), while pups of low-licking mothers grew up anxious. Over beers in Madrid with McGill colleague Moshe Szyf, an expert on chemical tags on DNA, the idea took shape. Meaney cross-fostered pups at birth: pups took on the profile of the mother who raised them. Licking altered methylation of the stress-hormone (glucocorticoid) receptor gene in the hippocampus, an epigenetic effect. In human brain tissue from suicide victims abused as children, the same receptor gene showed similar changes.
Behavioural epigenetics shows nature and nurture interact; the profile is mostly set by early childhood but can be changed with effort.
The inventories
Mlodinow reprints validated questionnaires so readers can map their profile:
| Inventory | Range | Typical (middle half) |
|---|---|---|
| Shame (11 scenarios) | 11 to 55 | 25 to 33 |
| Guilt (same scenarios) | 11 to 55 | 42 to 50 |
| Anxiety (20 items) | 20 to 80 | 31 to 39 (average 35; depressed patients 40s to 50s) |
| Anger (7 items) | 7 to 35 | 13 to 21 (average 17) |
| Verbal aggression (5 items) | 5 to 25 | 12 to 18 (average 15) |
| Oxford Happiness (29 items) | 29 to 174 | 95 to 135 (average about 115) |
| Romantic love/attachment (18 items) | 18 to 126 | 78 to 106 (average 91.5) |
- Shame versus guilt. Shame says "I am bad" and leads to hiding; guilt says "I did a bad thing" and leads to repair. Guilt is the healthier tendency. Shame emerges around age two; guilt needs more cognitive skill and appears later. Women score a couple of points higher on both.
- Anxiety answers potential, unpredictable threats (fear answers imminent ones). Too much is unhealthy, but some anxious people live longer because they take fewer risks.
- Anger and aggression evolved to defend access to food, water and mates; anger looks disproportionate because the cost of backing down could be survival. Remedies: leave the situation, and reappraise.
- Happiness has a genetic set point; we overrate how much money, cars or a team's win will lift us, but behaviour can raise it.
- Love floods the brain with dopamine (wanting) and lowers activity in areas for judgement and fear. Romantic love is thought to have evolved about 1.8 million years ago for pair bonding, and anthropologists found evidence of it in 147 cultures, including Tanzania's Hadza.
Jim and his shame
Jim cannot shake the sense that he is defective, not that he did a defective thing. Chapter 8 uses him to open the most practical distinction in the book: shame ("I am bad") versus guilt ("I did a bad thing"). Shame predicts hiding and self-attack; guilt predicts repair. Most people score higher on guilt than shame on the book's inventories; Jim's profile inverts that. Anxiety (potential, unpredictable threats) and fear (imminent threats) are another pair that feel similar and behave differently. Some anxious people, paradoxically, live longer in certain cohorts because vigilance catches real risks; the cost is lived experience.
| Shame | Guilt | |
|---|---|---|
| Focus | Global self | Specific act |
| Action tendency | Hide, attack self | Repair, apologise |
| Social function | Status / belonging threat | Norm repair |
| Clinical lean | More toxic when chronic | More adaptive when proportionate |
| Parenting cue | "What is wrong with you?" | "That choice hurt; fix it" |
Nature and nurture: Meaney's rats
Michael Meaney's lab showed that high-licking rat mothers raise pups with mellower stress profiles; low-licking mothers raise anxious ones. Cross-fostering flips the outcome toward the raising mother. The mechanism includes epigenetic methylation of a stress-hormone receptor gene. Parallel marks appear in some studies of abused human suicide victims; human evidence is messier and smaller. The book's honest claim: early care shapes the thermostat of the stress system; genes set ranges; profiles are sticky but not mystical. Mlodinow's mother and father are the human parable the rat data makes thinkable.
The inventories
Chapter 8 includes self-assessment questionnaires (summarised here by scoring ranges, not reprinted). Typical patterns: guilt scores sit high and shame low for most readers; anxiety and fear separate; anger and hostility vary widely. The point of an inventory is not a label for life; it is a map of defaults so that regulation (Chapter 9) can be chosen to fit the profile. A high-shame profile needs different work than a high-anger one. A high-anxiety profile may need interoceptive quieting before reappraisal will stick.
Profiles are sticky
His mother mellowed at 97 but "at her core she never changed." That sentence is the emotional realism of the book. Therapy, practice and relationships shift the band; they rarely delete it. That is not fatalism. It is why acceptance (Chapter 9) sits beside change.
- Epigenetic findings in rodents remain clearer than human parallels; treat human methylation claims as promising, not settled.
- Shame versus guilt distinctions continue to guide clinical work; cultural variation in which emotion is named "shame" matters.
- Self-report inventories are useful maps and poor destinies; combine with behaviour over weeks.
SHAME HIDES; GUILT REPAIRS. Anxiety watches the maybe; fear faces the now. Genes and licking set the thermostat; inventories read it.
Profile talk becomes identity prison ("I'm just an anxious person") or ranking ("guilt people are healthier than shame people" as moral worth). Shadow: using Meaney to blame mothers for adult anxiety, or using inventories to excuse harm.
Profiles expanded: shame, care and sticky thermostats
Jim's sentence
"I am bad" versus "I did bad" is a small linguistic difference with large behavioural consequences. Tangney and others built careers on that distinction. Parenting, management and self-talk either install shame ("what is wrong with you?") or guilt ("that choice hurt; fix it"). Cultures differ in how much global self is staked on a single act. Immigration, religion and class all shape the mix. Jim is not a freak; he is a profile peak that many readers will recognise in quieter form.
Anxiety's double face
Trait anxiety predicts worry and avoidance; in some epidemiological pockets it also predicts catching real risks earlier. The book's refusal to treat anxiety as pure defect echoes Chapter 1's melanoma story. Regulation aims to keep the vigilance and lose the paralysis. Fear, by contrast, is for the tiger in the room; treating every maybe-threat as a now-threat is the anxiety error.
Meaney, mothers and misuse
Epigenetic care effects in rats are among the cleanest findings in behavioural neuroscience. Human translation is correlational, ethically constrained and noisy. The misuse pattern is predictable: journalists blame mothers; politicians cut supports while preaching resilience; adults weaponise the finding in family fights. The right use is policy and compassion: early care is public infrastructure for emotional thermostats, and adults with harsh thermostats deserve tools (Chapter 9), not lectures about their mothers.
Inventories as temporary maps
Scores shift with mood, sleep and recent events. Retake after a month of practice before treating a number as identity. Combine self-report with a behaviour log: what did shame make you hide this week? What did guilt make you repair? Profiles earn their keep when they change choices, not when they decorate a journal.
Stickiness and hope
"At her core she never changed" is hard to hear in a self-help culture. It is also what makes acceptance coherent. If profiles were infinitely plastic, acceptance would be laziness. Because bands are real, working within a band is wisdom. Therapy, meds, relationships and practice move bands; they rarely mint a new person. That is enough hope for an honest book.
Cross-fostering and the refusal of destiny
Meaney's cross-fostering result is the anti-destiny finding: the raising mother's care can override the birth mother's statistical legacy in rats. Human lives are not cross-fostering experiments. The ethical reading is not "parents caused your shame score"; it is "profiles are built, therefore they can be worked with." Inventories are mirrors for that work, not verdicts for a court.
- Inventory sources. Shame and guilt come from Tangney's TOSCA; anxiety from Spielberger's trait anxiety scale (STAI); anger and aggression from the Buss-Perry questionnaire; happiness from Hills and Argyle (2002); attachment from an adult attachment scale. They are research tools; self-scoring at home gives a rough picture, not a diagnosis.
- Human epigenetics. Meaney's rat work is strong; human findings rest on small post-mortem samples and correlational blood studies that have replicated inconsistently. Claims of epigenetic inheritance of trauma across generations in humans remain contested.
- The romantic-love survey (Jankowiak and Fischer, 1992) found it in 147 of 166 cultures; the 1.8-million-year date is speculative.
- "Set point from your DNA": twin studies put happiness heritability at roughly 30 to 50%; Lyubomirsky's popular "40% under your control" pie chart has been criticised as a misreading.
- Affective style research (Davidson's "emotional style") and ecological momentary assessment now track emotional profiles in daily life by phone, revealing that emotional inertia (slow recovery) predicts depression risk.
- Large genome-wide studies find thousands of variants each with tiny effects on neuroticism and well-being; no single "anxiety gene."
- Attachment theory in adults is well supported, but popular "attachment style" labels online often overstate how fixed they are; styles shift with relationships and therapy.
- The World Happiness Report 2026 continues to show Nordic countries at the top and declining well-being among young people in North America.
Score yourself on one or two of the inventories, then ask someone who knows you well to score you as they see you. Differences are informative: we often misjudge our own profile, as Jim did. Look at each emotion's four parameters. If anger has a low threshold and fast peak, build in pauses; if sadness has a long recovery, plan supportive routines after setbacks. If your shame score is high relative to guilt, practise separating "I did something bad" from "I am bad." Remember that there is no ideal profile: some anxious people use their anxiety productively, and the aim is awareness, not uniformity.
Each of us has an emotional profile: threshold, speed, intensity and recovery for each emotion. It is shaped by genes and, powerfully, by early care (Meaney's licking rats). Validated inventories for shame, guilt, anxiety, anger, happiness and love let you see your own fingerprint, the first step toward managing it.
9. Managing Emotions
Contagion
In October 2011 a cheerleader at Le Roy High School woke with uncontrollable facial twitching. Within weeks her best friend and then more than a dozen other students, almost all adolescent girls, had similar tics. Experts suspected a post-strep immune reaction (PANDAS) or toxic waste; nothing was found. The diagnosis: mass psychogenic illness, which is commoner than people think (North Carolina in 2002, and elsewhere).
Adam Smith (1759) noted that we flinch when we see a blow aimed at someone else. Studies of emotional contagion have risen tenfold. A colleague who habitually complains can leave you anxious after a chat even though you felt fine before. Mimicry evolved because synchronising emotions helped our ancestors bond and coordinate in groups, as primates still do. We unconsciously mimic faces, voices and postures, sometimes faster than conscious reaction allows, and our feelings follow. A Harvard and UC San Diego study of 4,739 people over 20 years found happiness spreads through social networks. A secret Facebook experiment showed that tweaking users' feeds shifted the emotional tone of their own posts. The hopeful corollary: our smiles spread too.
Why regulate?
Emotions drive our goals but can also derail them. Regulation is especially human; animals share our neurotransmitters, but we can reflect on feelings. Good regulators have better cardiac health; in a 13-year study of older men, those least able to regulate emotion had the most heart disease. Only in the last decade or two have researchers rigorously tested which methods work.
Acceptance: the power of Stoicism
In September 1965 naval wing commander James Stockdale's A-4 Skyhawk was hit over North Vietnam. During the 30-second parachute ride he told himself: "Five years down there at least. I'm leaving the world of technology and entering the world of Epictetus." Beaten on landing (his leg was crippled for life) and tortured for seven and a half years in the "Hanoi Hilton," he became a rock for other prisoners. At Stanford he had been given Epictetus's Enchiridion, which stayed on his bedside table through three combat tours.
Stoicism does not reject comfort; it warns against depending on what you cannot control. Epictetus, himself a former slave, wrote: "A man's master is he who is able to confirm or remove whatever that man seeks or shuns." If your well-being depends only on what is in your power (your judgements, choices and efforts), no captor can master you. Tying your peace of mind to another person's behaviour is as futile as tying it to the weather; of anything outside your control, Epictetus advised, "be prepared to say that it is nothing to you." In the camp, Stockdale stopped worrying about release dates and concentrated on what he could do each day: communicate with fellow prisoners by tapping codes, resist in small ways, and keep his self-respect. "It's not our circumstances that get us down but rather the judgments we make about them." Acceptance means acknowledging that the worst may happen and focusing on what you can do. In a lab game, students taught acceptance kept playing significantly longer despite painful electric shocks.
Reappraisal: the power of spin
Twenty minutes late after a botched detour, you can think "Why can't those idiots..." or "a chance to listen to more music." Appraisal, deciding what an event means, is a phase of every emotion, and reappraisal changes the meaning and so the feeling.
- Runners. At the US Army's Natick Soldier Systems Center, 24 young people ran on a treadmill; those coached to reappraise discomfort (for example, as a sign of getting fitter) reported less negative feeling and ran with less perceived effort.
- Traders. Mark Fenton-O'Creevy (Open University), a former groundskeeper, chef, maths teacher and therapist, interviewed 118 traders and ten managers at four investment banks. The least successful traders tried to suppress or ignore emotion and were swamped by it. The best acknowledged emotions, reflected on them and used reappraisal ("big losses happen"). One manager: "I have to play the role of director of emotion."
Kahneman's System 1 (fast, unconscious) does most trading; emotion is its "radar," as disgust at a wormy oyster saves analysis. Regulation keeps the radar from running amok.
Expression: the power of words
Karen S., COO of a Hollywood production company, vents to a trusted colleague after hard meetings. Most people think talking amplifies negative feelings; research says it defuses them, especially with sympathetic listeners. Researchers call it affect labelling: naming a feeling lowers distress after disturbing images, eases public-speaking anxiety and reduces fear of spiders. After a taxi totalled his car at a red light, Mlodinow's retelling eased the shock.
Researchers scanned more than a billion tweets from 600,000 users for explicit statements ("I feel sad") and analysed the surrounding 12 hours from about 100,000 people with the VADER sentiment program. Negative emotion built quickly in the hour before the statement and then dropped sharply afterwards. Positive emotion built and faded slowly. Putting a feeling into words calms it. As Shakespeare's Malcolm says: "Give sorrow words."
The joy of emotion
His mother warned that a bad reputation is hard to shed. Emotion has had a bad reputation for 2,000 years; Mlodinow hopes he has shed it. Humans thrive in more ecosystems than any species because emotion lets us respond flexibly. Know your profile, accept, reappraise and express.
Contagion at Le Roy and in ordinary life
In October 2011 a cheerleader at Le Roy High School woke with facial twitching and jaw spasms. Her best friend followed; then more than a dozen students, almost all adolescent girls. Neurologists considered PANDAS (post-strep autoimmune) and toxic waste; nothing was found. The diagnosis was mass psychogenic illness, cousin to the 2002 North Carolina cluster. Emotion here spreads as symptom. Adam Smith (1759) already noted that we flinch when a blow is aimed at someone else's limb. Modern work multiplies the point: mimicry of faces, voices and postures can run faster than conscious appraisal; feelings follow the face. A Harvard and UC San Diego study of 4,739 people over 20 years found happiness spreading through social networks. A secret Facebook experiment shifted the emotional tone of users' posts by tweaking feeds. Contagion is not only pathology; it is how groups synchronise. The hopeful corollary stands: smiles spread too. The A closer look also stands: network contagion studies are sensitive to modelling choices; treat magnitudes cautiously.
Why regulate?
Emotions set goals and can derail them. Regulation is especially human: we share neurotransmitters with other animals, but we can reflect on feelings and choose strategies. Good regulators show better cardiac health in long cohorts; a 13-year study of older men linked poor regulation to more heart disease. James Gross's process model places interventions along the sequence: situation selection and modification, attentional deployment, cognitive appraisal, then response modulation. Mlodinow's three recommended tools (acceptance, reappraisal, expression) act mainly on appraisal and felt emotion. Suppression (hiding the face) acts late and costs more: raised blood pressure, worse memory for the conversation, leaked emotion that partners still detect.
Acceptance: Stockdale and Epictetus
In September 1965 naval wing commander James Stockdale's A-4 was hit over North Vietnam. In the thirty-second parachute ride he told himself he was leaving technology for the world of Epictetus. Tortured for seven and a half years in the Hanoi Hilton, leg crippled from the landing beating, he became a rock for other prisoners. At Stanford he had been given the Enchiridion; it stayed by his bed through three combat tours. Stoicism here is not grim denial of comfort; it is refusing to hitch your peace to what you cannot control. Epictetus, a former slave: your master is whoever can grant or remove what you seek. Acceptance means naming that the worst may happen and focusing effort on judgements, choices and daily acts (tap codes, small resistance, self-respect). In lab games, students taught acceptance kept playing significantly longer under painful shocks. Stockdale's line, which the book echoes: circumstances do not get us down so much as the judgements we make about them.
Reappraisal: runners and traders
Reappraisal changes the meaning of an event and so the feeling. Late for a meeting, you can curse "those idiots" or hear more of a podcast; the body state shifts with the story. At the US Army Natick labs, 24 young runners coached to reappraise treadmill discomfort as a sign of getting fitter reported less negative feeling and less perceived effort. Mark Fenton-O'Creevy interviewed 118 traders and ten managers at four investment banks. The least successful tried to suppress or ignore emotion and were swamped by it. The best acknowledged emotion, reflected, and reappraised ("big losses happen"). One manager called himself "director of emotion." That phrase is the chapter's enterprise takeaway: regulation is a management skill, not a private luxury.
Expression: the power of words
Putting a feeling into words changes its course. A study of over a billion tweets (described in the book) found negative emotion building in the hour before "I feel sad" and dropping sharply after; positive emotion rose and faded more slowly. Labelling recruits prefrontal regions that damp amygdala reactivity in many paradigms. Expression is not venting without bound; chronic rumination without reappraisal can worsen mood. The useful form is name, then choose: label precisely (granularity from Chapter 5), then accept or reappraise, rather than rehearse the injury.
| Tool | Where it acts | Best for | Failure mode |
|---|---|---|---|
| Acceptance | Felt emotion + judgement | Pain you cannot change now | Passive resignation that skips agency |
| Reappraisal | Meaning of the event | Ambiguous setbacks, effort, loss | Toxic positivity; denying real harm |
| Expression / labelling | Bringing feeling into words | Overwhelm, unclear affect | Rumination loops; oversharing as substitute for action |
| (Avoid) Suppression | Response only | Short social necessity | Costly if chronic; partners still read the leak |
- Meta-analyses through 2025 support reappraisal and acceptance with culture as a moderator: which strategy fits depends partly on cultural norms about emotion.
- Contagion and network-emotion findings remain directionally robust; effect sizes and causal identification stay debated.
- Gross's process model is still the teaching scaffold; newer work adds interpersonal and cultural layers.
ACCEPT → REAPPRAISE → NAME. Let it be, change the story, put it in words. Contagion means your tools (and your mood) are partly public goods.
Stoicism becomes emotional austerity that shames tears. Reappraisal becomes spin that protects abusers. Labelling becomes intellectualising that never feels. Shadow: using Chapter 9 to silence yourself or others, rather than to choose a tool that fits the profile from Chapter 8.
The joy of emotion
The chapter closes by refusing the old dream of a life without feeling. Emotion is how meaning shows up in a body. The aim is not numbness; it is skill. That sentence is the bridge into the Epilogue, where skill meets a death the author cannot reappraise away.
Regulation expanded: contagion, tools, institutions
Contagion as design input
Open offices, group chats and trading floors are contagion machines. Le Roy is the pathological extreme; the Monday morning mood dump is the common form. Leaders who ignore contagion treat culture as slogan; leaders who respect it manage arrivals, meeting tone and who speaks first after bad news. Smith's flinch is the micro-basis; the 4,739-person network study is the macro-hint; Facebook's feed experiment is the platform warning.
Acceptance is not passivity
Stockdale's acceptance included tap codes, resistance and leadership. Epictetus separates control from non-control so that effort concentrates. Corporate "acceptance" that means "stop complaining about overload" is a counterfeit. The test: after acceptance, does agency rise on the controllable slice? If not, you have resignation, not Stoicism.
Reappraisal's integrity constraint
Reappraisal that denies harm is gaslighting, including self-gaslighting. Reappraisal that finds a usable frame ("this loss is information"; "this pain marks adaptation") preserves truth and changes feeling. Natick runners and Fenton-O'Creevy's best traders stay on the honest side of that line. The least successful traders' suppression is the foil: emotion ignored becomes emotion in charge.
Expression versus rumination
Labelling that ends in a plan or a release differs from labelling that loops. Journaling protocols with time limits, or speaking once to a trusted person with a request, beat endless group-chat venting. The tweet-study shape (negative affect falling after naming) is the promise; clinical rumination research is the warning.
Cardiac health and the long game
The 13-year heart-disease link for poor regulators is a reminder that Chapter 9 is public health, not only mood. Suppression's blood-pressure cost is acute; decades of mode mismatch are chronic. Regulation skill is closer to brushing teeth than to a weekend workshop.
Institutions as directors of emotion
"Director of emotion" should be a recognised function in high-stakes teams: who sets the mode for the incident? Who cools contagion after a loss? Who prevents shame from freezing the postmortem? Aviation's crew resource management already does versions of this. Finance and tech often pretend affect is unprofessional and then act out the unprofessional affect anyway.
Director of emotion
Fenton-O'Creevy's trader-manager who called himself director of emotion names the institutional version of Gross's model. Teams have contagion (Le Roy in mild form). Cultures have display rules. Leaders who suppress and demand suppression pay in blood pressure and in bad trades. Leaders who accept, reappraise and label build rooms where information about risk can surface.
- The "secret Facebook experiment" (Kramer and colleagues, 2014, about 689,000 users) produced a tiny effect, about one word per thousand, and caused an ethics controversy over consent.
- Christakis and Fowler's happiness network results (the 4,739-person Framingham data) were criticised for being unable to separate contagion from shared environments and similar friends.
- The Twitter study (Fan and colleagues, 2019) appeared in Nature Human Behaviour, not Nature, and is observational: people may tweet "I feel sad" once the worst has passed.
- Mass psychogenic illness was the diagnosis at Le Roy, but some families disputed it; it remains a diagnosis of exclusion.
- Stockdale is also the source of Jim Collins's "Stockdale paradox" (confront brutal facts while keeping faith that you will prevail).
- Reappraisal remains the most studied and generally effective strategy; meta-analyses also show context matters: reappraisal helps most with uncontrollable stress, while problem-solving helps when change is possible ("regulatory flexibility," Bonanno). Culture moderates strategy fit (2025 meta-analyses).
- Suppression reliably impairs memory and social connection, as Gross's lab first showed.
- Affect labelling has replicated in fMRI work (lower amygdala activity) and in field studies; expressive writing (Pennebaker) has small but positive average effects.
- Acceptance and Commitment Therapy (ACT) has a broad evidence base and draws directly on the acceptance idea.
- Social media contagion: large studies through 2025 find small average links between social media use and teenage mental health, with heated debate over causation (Haidt versus Odgers and others). The US Surgeon General called in 2024 for warning labels on social media.
- AI companions and chatbots used for emotional support are now common; early trials of therapy chatbots show short-term symptom relief, and regulators and researchers warn about dependence and safety.
A simple sequence for difficult moments: notice the feeling in your body; name it in specific words ("I feel humiliated and anxious"); accept it as information rather than fighting it; reappraise by asking what else this situation could mean and what is in your control; then express it to a trusted person or in writing. Avoid habitual suppression. Guard your emotional inputs, since contagion works through people, news and feeds.
Emotions spread from person to person, so manage your inputs. Three tested tools: acceptance (Stockdale's Stoicism), reappraisal (the best traders), and expression or labelling (give sorrow words). Suppression is costly. Contagion means your regulation is partly a public good.
Epilogue: The Goodbye
In 2020 COVID-19 swept his mother's assisted living home. At 98, with dementia, she fell gravely ill. Her doctor needed a decision by 6:00 p.m., eight minutes away: send her to the hospital, which she hated, or let her die peacefully at home. Mlodinow, "the most even-keeled, unflappable person" his son knew, collapsed in tears and could not think. Remembering the traders, he accepted his emotion instead of fighting it. He had decided on home, yet when the doctor answered he heard himself say "send her to the hospital." Like his father at the fence, he decided one thing and did another.
She did well at first; he spoke to her over FaceTime on the ward's single iPhone. Her home, overwhelmed, kept delaying her return. Then she declined and died a little after 3 a.m. He does not regret listening to his heart. It still hurts to imagine her dying without her family around her, but he believes the hospital gave her a real chance, and that the decision came from the integrated judgement the book describes: years of experience, love and fear combined faster than conscious analysis could manage. The episode is the book's argument in miniature. Emotion did not stop him thinking; once accepted rather than resisted, it was his thinking. Regulation did not mean feeling less, but letting feeling do its job. Managing emotion "is not just a science but an art"; his friend Deepak Chopra seems to meet any news with equanimity.
What Hitler could not do
For years his mother, wheelchair-bound but content in assisted living, received weekly visits. Then COVID-19 entered the home. Staff and residents fell ill. At ninety-eight, with some dementia, the decision fell to her son under an eight-minute clock.
Plan versus act
She considered hospitals torture: strange rooms, hated IVs and catheters, strangers in and out, absence of known caregivers. His prior plan was home. His spoken act, after accepting the flood of feeling, was hospital. She gained a chance; she still died after 3 a.m., family not at the bedside in the way he wished. He does not tidy the contradiction into a slogan. The book's tools (acceptance, reappraisal, expression) meet their limit here. You cannot reappraise away a mother's death. You can accept the feeling, act from the integrated judgement, and live with the hurt that remains.
The expert is overwhelmed
The epilogue's intellectual job is to prevent the rest of the book from sounding like a mastery fantasy. Affective neuroscience explains mechanisms; it does not exempt the explainer. Contagion, profile stickiness, wanting and determination all still operate when the person who wrote the chapters is the one grieving. The introduction's question (why two survivors, two emotional lives) returns as a farewell: her anxious profile shaped decades of family life; it also shaped a long survival. He does not resolve that into a neat moral. He says goodbye.
Unfinished profiles
Anxious vigilance is costly and sometimes protective; optimistic confidence is buoyant and sometimes blind. Two survivors, two tools, one history. The book began with that unsolved comparison and ends without solving it, which is a form of respect. Technique bows to attachment: use acceptance, reappraisal and naming when they fit; put them down when love needs only presence. Do not turn this goodbye into a checklist for other families' ICU decisions, or Stockdale into a demand that grief be tidy, or wanting versus liking into a party game over a wake.
TOOLS MEET LOVE. Accept the feeling; let integrated judgement act; reappraisal cannot erase the loss. The book ends where skill bows to attachment.
Write, for someone you love, the one-paragraph preference you would want honoured if you held a similar decision: setting, people, what to refuse, what chance you would still take. Share it if you can. That is the epilogue's exercise without waiting for a crisis clock.
Using the epilogue to demand that every death be "accepted" beautifully, or to judge families who choose aggressive care (or who refuse it), inverts the humility of the chapter. Shadow: turning a son's goodbye into a rule for strangers.
Even an expert can be overwhelmed. Mlodinow's final decision about his mother came from accepting his emotion, not suppressing it; plan and spoken act diverged, as Simon's pause once did. The science of emotion does not cancel grief. Emotion is not the enemy of clear action; it is often what makes the action matter.
Synthesis
25 takeaways
- Emotion is a tool of reason, not its enemy; we cannot decide without it.
- The traditional view (few basic emotions, fixed triggers, limbic seat) is at best a rough approximation.
- The triune brain is a myth; emotion emerges from interacting networks.
- Each emotion is a mode of operation that resets goals, attention and weights.
- Anxiety produces a pessimistic bias toward ambiguous data, even in bees.
- Emotions differ from reflexes in valence, persistence, generalisability, scalability and automaticity.
- Emotion buys a delay between trigger and response, letting reason in.
- Core affect (valence and arousal) is the body's constant report to the brain.
- Hunger, cold and fatigue colour judgements without our noticing (juncos gamble when cold).
- The gut and its microbes talk to the brain.
- The brain is a prediction machine; feelings are part of the forecast.
- Fear inflates risk; anger shrinks it; disgust drives rejection.
- Positive emotions broaden thinking; sadness drives re-evaluation.
- Feelings are constructed from body, context, experience and culture.
- The same bodily state can feel like pain, relief, euphoria or anger.
- Emotion words carve experience differently across cultures.
- Wanting (dopamine) and liking (opioids) are separate systems.
- Addiction and supernormal stimuli inflate wanting while liking fades.
- Determination comes from salience plus executive networks; exercise, mindfulness and sleep strengthen it.
- Everyone has an emotional profile: threshold, latency, intensity, recovery.
- Early care shapes stress responses (Meaney's rats).
- Guilt ("I did bad") is healthier than shame ("I am bad").
- Emotions are contagious; choose your inputs.
- Acceptance, reappraisal and labelling work; suppression costs.
- Self-knowledge is the first step: notice the feeling, name it, then choose.
Frameworks to reuse
| Framework | Use it when | Ch |
|---|---|---|
| Five properties of emotion states | Deciding whether behaviour reflects a state or a reflex | 2 |
| Core affect plane | Checking whether hunger or fatigue is driving a decision | 3 |
| Emotion as mode of operation | Asking which emotion is steering a judgement | 4 |
| Construction model | Reinterpreting arousal (nerves as excitement) | 5 |
| Wanting versus liking | Spotting cravings that no longer satisfy | 6 |
| Salience plus executive | Building persistence on a long goal | 7 |
| Profile parameters | Understanding your own and others' reactions | 8 |
| Process model of regulation | Picking a strategy before, during or after an emotion | 9 |
The book chapter by chapter
| Ch | One-line summary |
|---|---|
| Intro | Two survivors, two emotional styles: what makes us feel as we do? |
| 1 | From Plato to Petrov: emotion is a tool of thought, not its saboteur |
| 2 | Emotions are flexible, persistent, scalable states that improve on reflexes |
| 3 | Core affect links body and mind and silently biases our choices |
| 4 | Each emotion is a distinct mode of operation for the brain |
| 5 | Feelings are constructed from body, context, experience and culture |
| 6 | Wanting and liking are separate systems, and modern products exploit the gap |
| 7 | Determination comes from salience and executive networks and can be trained |
| 8 | Everyone has a measurable emotional profile shaped by genes and early care |
| 9 | Acceptance, reappraisal and expression are the regulation tools that work |
| Epilogue | Even an expert can be overwhelmed; accepting emotion guided his last decision for his mother |
Questions for discussion
- Mlodinow's parents endured similar horrors yet reacted differently. Which parts of the book best explain the difference: genes, early experience, core affect or regulation?
- Was Petrov's choice driven by emotion, by reasoning, or by both?
- Anderson and Adolphs define emotion states without reference to feelings. Is that enough to say a fruit fly has emotions?
- If feelings are constructed, can you choose to feel differently? Where are the limits?
- Which products or services in your life are designed to inflate wanting more than liking?
- Given the critiques of the hungry-judges study, what safeguards would you still recommend for people who make many decisions in a row?
- Which of the three regulation strategies do you use most, and which would serve you better?
- Should AI systems ever try to detect or simulate emotion? What does the EU's ban on workplace emotion recognition suggest?
A closer look
- Replication. Several headline studies (hungry judges, disgust and morality, Schachter-Singer, arousal and judgement, small mindfulness trials) are weaker than presented. The core claims survive; some vivid illustrations do not.
- Partisan on theory. Mlodinow co-wrote with Lisa Feldman Barrett and leans constructionist, while still using basic-emotion language. The 2025 literature treats the theories as partly complementary.
- Animal to human leaps. Bees, flies, ducks, juncos and mice show emotion states; the book sometimes implies feelings.
- Self-help inventories are valid research instruments but self-scoring gives only a rough picture.
- Strengths: clear writing, vivid personal stories, and a sound core message (emotion and cognition are integrated) that is mainstream in 2026.
Who's who
| Person | Role |
|---|---|
| Ralph Adolphs, David Anderson | Caltech; five properties of emotion |
| Lisa Feldman Barrett | Theory of constructed emotion |
| Kent Berridge | Wanting versus liking |
| Antonio Damasio | Emotion, body and decision making |
| Charles Darwin | Expression of the Emotions (1872) |
| Mark Fenton-O'Creevy | Emotion in securities traders |
| Barbara Fredrickson | Broaden-and-build theory |
| Adam Galinsky | Perspective-taking in negotiation |
| Jonathan Haidt | Moral intuition and disgust |
| Robert Heath | Deep brain stimulation, "pleasure centre" |
| Michael Meaney | Maternal care and epigenetics |
| James Olds, Peter Milner | Self-stimulating rats (1954) |
| Josef Parvizi | Stimulated "determination" (2013) |
| Stanislav Petrov | Averted nuclear false alarm (1983) |
| Peter Salovey, John Mayer | "Emotional intelligence" (1990) |
| Stanley Schachter, Jerome Singer | Arousal plus interpretation (1962) |
| James Stockdale | Stoic POW |
| Niko Tinbergen | Supernormal stimuli |
Study aids index
| Ch | TO REMEMBER | Shadow risk |
|---|---|---|
| Intro | TWO SURVIVORS → ONE QUESTION | Ranking resilient vs broken |
| 1 | ENTERPRISE → BEES → PETROV | Blank-cheque impulsivity |
| 2 | REFLEX → SCRIPT → BUTTON → EMOTION | Dismissing present harm as baggage |
| 3 | BODY → PREDICT → AFFECT → FEELING | Body-as-oracle or body-as-irrelevant |
| 4 | MODE, NOT NOISE | Weaponised modes; toxic positivity |
| 5 | AROUSAL + CONTEXT + CONCEPT | "Emotions aren't real" gaslight |
| 6 | WANT ≠ LIKE | Dopamine moral panic |
| 7 | SALIENCE + EXECUTIVE + MEANING | Grind culture |
| 8 | SHAME HIDES; GUILT REPAIRS | Profile as prison; mother-blame |
| 9 | ACCEPT → REAPPRAISE → NAME | Tools used to silence |
| Epilogue | TOOLS MEET LOVE | Policing others' grief |
Since publicationfield notes
- Constructed vs basic emotion: still incommensurate; teach both; Barrett replies and universality critiques both active.
- Hungry judges: overstated; keep interoception lesson; drop the cartoon graph as law.
- Incentive sensitisation: 30-year review (2025) supports wanting/liking dissociation; opioid work continues through 2026.
- Emotion regulation meta-analyses (2025): reappraisal and acceptance work; culture moderates.
- Triune brain: neuromyth label sticks in 2025 reviews; popular training lags.
Glossary
| Term | Meaning |
|---|---|
| Affective neuroscience | Study of the brain mechanisms of emotion |
| Affect labelling | Putting feelings into words, which reduces their intensity |
| Allostasis | The brain anticipating and meeting the body's needs before they arise |
| Amygdala | Almond-shaped structure linked to arousal, threat and salience |
| Anterior cingulate cortex | Midline region in the salience network; conflict and effort |
| Appraisal | Interpreting what a situation means for you |
| Arousal | The energised-to-sleepy dimension of core affect |
| Basic emotion theory | View that a few emotions are innate, universal and distinct |
| Broaden-and-build | Fredrickson's theory that positive emotions widen thinking and build resources |
| Constructed emotion | Barrett's view that emotions are built from core affect, context and concepts |
| Term | Meaning |
|---|---|
| Core affect | Continuous two-dimensional state of valence and arousal |
| Determination | Firmness of purpose in pursuing a goal |
| Drive theory | Old view that we act mainly to reduce discomfort |
| Emotional contagion | Spread of emotion between people, often via unconscious mimicry |
| Emotional intelligence | Ability to perceive, understand, use and manage emotions |
| Emotional profile | Personal pattern of threshold, latency, intensity and recovery for each emotion |
| Enteric nervous system | The gut's own nerve network, the "second brain" |
| Epigenetics | Changes in gene activity, such as methylation, without changes in DNA sequence |
| Executive control network | Prefrontal and parietal regions that keep you on task |
| Gut-brain axis | Two-way communication between gut and brain |
| Term | Meaning |
|---|---|
| Hedonic hotspot | Small brain site where opioids amplify liking |
| Homeostasis | Keeping bodily variables within safe limits |
| Hyperpalatable food | Engineered food that exceeds natural reward |
| Insula | Cortex that integrates body signals; key to interoception |
| Interoception | Sensing the internal state of the body |
| Liking | Pleasure itself, mediated by opioids and endocannabinoids |
| Mass psychogenic illness | Physical symptoms spreading through a group without organic cause |
| Nucleus accumbens | Hub of the wanting system |
| Optogenetics | Controlling neurons with light |
| Orbitofrontal cortex | Prefrontal area linked to value and valence |
| Term | Meaning |
|---|---|
| Placebo effect | Improvement from belief in a treatment |
| Production rule system | Model of thought as a set of if-then rules |
| Reappraisal | Changing the meaning of a situation to change the feeling |
| Salience network | Network that flags what matters in body and world |
| Shame / guilt | Negative judgement of the self / of a specific action |
| Supernormal stimulus | Artificial stimulus stronger than any natural one |
| Suppression | Hiding the outward expression of an emotion |
| Triune brain | Discredited model of reptilian, limbic and neocortical layers |
| Valence | The pleasant-to-unpleasant dimension of feeling |
| Wanting | Incentive salience, mediated by dopamine |
| Allostasis | Anticipatory physiological regulation; broader than fixed-set-point homeostasis |
| Incentive salience | Berridge's term for dopamine-related wanting pull toward cues |
| Emotional granularity | Ability to differentiate fine-grained emotion concepts |
| Mass psychogenic illness | Contagion of symptoms without organic toxin; Le Roy 2011 |
| Process model (Gross) | Situation → attention → appraisal → response; where regulation intervenes |
About this book
Emotional: How Feelings Shape Our Thinking by Leonard Mlodinow, 2022. This is a chapter-by-chapter study of the book, with summaries, maps, tables and diagrams; short quotations keep their page references. It is shared free for personal study, is not for sale, and is not endorsed by the author or publisher. Please read and buy the original.
Insight Editions · rajeshrmahapatra.com/insight-editions