Case Study 1 โ€” Theo's 21:40: Diagnosing a Cue Instead of a Character

โš ๏ธ Theo Vasquez is the book's anchor case, introduced in Chapter 1 and an illustrative composite throughout. His numbers โ€” RMR 1,908, TDEE 2,670, intake 3,160 kcal, fiber 14 g, added sugar 78 g, A1c 5.9% โ€” have been canon since Chapter 4. โš ๏ธ This case study returns to the one behaviour that none of the preceding thirty-two chapters fixed.**


Setup

โš ๏ธ By March, Theo had done nearly everything this book recommends.

โš ๏ธ Status
Fibre โš ๏ธ 14 g โ†’ 29 g (Ch 11)
Added sugar โš ๏ธ 78 g โ†’ ~35 g (Ch 18)
Alcohol โš ๏ธ 6% of intake โ†’ ~2% (Ch 12)
Labels โš ๏ธ He can decode a shelf in nine minutes (Ch 30)
โš ๏ธ Meal planning โš ๏ธ Nine slots, components on Sunday, bad-night meal named (Ch 31)
A1c โš ๏ธ 5.9% โ†’ 5.6%
โš ๏ธ The 21:40 problem โš ๏ธ Unchanged. Four to five nights a week, for eleven years

โš ๏ธ His own account, and it is precise:

"I'm not hungry. I know I'm not hungry. I finish the washing up, I sit down, and about ten minutes later I'm in the kitchen. I've done this since university."

โš ๏ธ Everything he had been told to do about it:

โš ๏ธ Advice received โš ๏ธ Which system it addressed โš ๏ธ Outcome
"Just don't buy them" โš ๏ธ Hedonic โ€” cue removal โš ๏ธ He'd never actually tried it. See below
"Drink a glass of water first" โš ๏ธ Homeostatic โš ๏ธ Wrong system. No effect
"Wait twenty minutes and see if you still want it" โš ๏ธ Homeostatic โš ๏ธ He waited. He still wanted it. Twenty minutes later
โš ๏ธ "Ask yourself if you're really hungry" โš ๏ธ Neither โ€” a diagnostic that doesn't diagnose โš ๏ธ He always knew he wasn't. It changed nothing and made him feel worse
"More willpower" โš ๏ธ Nothing โš ๏ธ Eleven years of evidence

โš ๏ธ Four of the five addressed the system that wasn't involved.


The diagnosis

โš ๏ธ Two weeks of the ยง33's Project Checkpoint log. Four columns, no calories, no judgement.

โš ๏ธ Time โš ๏ธ Context โš ๏ธ What started it โš ๏ธ Wanting / liking
21:38 โš ๏ธ Sat down, TV on โš ๏ธ Finished washing up โš ๏ธ W 9 / L 5
21:45 Sofa โš ๏ธ Finished washing up โš ๏ธ W 8 / L 4
13:10 โš ๏ธ Desk, working โš ๏ธ Skipped lunch โ€” 6 h since food โš ๏ธ W 7 / L 8
21:52 Sofa โš ๏ธ Finished washing up โš ๏ธ W 9 / L 3
16:30 Office kitchen โš ๏ธ Someone brought cake W 6 / L 8
21:40 Sofa โš ๏ธ Finished washing up โš ๏ธ W 9 / L 4

โš ๏ธ Nine of eleven evening occasions had the same trigger, within a seven-minute window.

๐Ÿ’ก Aha moment. โš ๏ธ "It's not the evening. It's the washing up."

โš ๏ธ The cue was the END OF A TASK. โš ๏ธ The transition from doing something to doing nothing, which for eleven years had been marked by going to the kitchen โ€” a room he was already standing in.

โš ๏ธ And the wanting/liking columns are the second finding.

โš ๏ธ Evening occasions: wanting 8โ€“9, liking 3โ€“5. Daytime occasions: wanting 6โ€“7, liking 8.

โš ๏ธ The evening eating was almost pure WANTING. He genuinely did not enjoy it much, and had been treating that fact as evidence that he was irrational.

โš ๏ธ It is evidence of two systems. (ยง33.2.)


โš ๏ธ The intervention, and what was deliberately not done

โš ๏ธ Three changes. None of them was "try harder at 21:40."

โš ๏ธ 1. Break the cue

โš ๏ธ The washing up moved to before dinner where possible, and โ€” more importantly โ€” the transition acquired a different behaviour.

โš ๏ธ He made tea. Every night, immediately on finishing, without deciding.

โš ๏ธ This is ยง33.8's substitution: the cue still fires, and something else follows it. โš ๏ธ It is not a nutritional intervention and the tea is not doing anything metabolic. It occupies the slot.

2. โš ๏ธ Remove the object

โš ๏ธ The tortilla chips stopped entering the house.

โš ๏ธ His objection is worth recording because it is the standard one:

"That feels like admitting I can't control myself."

โš ๏ธ The reframe from ยง33.12 is what made it acceptable to him: you are not being asked to be stronger. You are making the decision once, on a Saturday, in a shop, unhungry โ€” where it is easy โ€” instead of eleven times a week at 21:40, where it is not.

โš ๏ธ "The people who look disciplined are usually not winning more fights. They are having fewer."

3. โš ๏ธ Fix the 13:10, which was a different problem entirely

โš ๏ธ The daytime occasions had a different signature: interval-initiated, high liking, six hours since food.

โš ๏ธ That is homeostatic, and it took a homeostatic tool: he ate lunch. โš ๏ธ Applying the cue intervention to it would have failed, and applying "drink water and wait" to the evening had already failed.

โš ๏ธ What was NOT done

โš ๏ธ Deliberately not attempted โš ๏ธ Why
โš ๏ธ A rule banning snacks โš ๏ธ ยง33.9 โ€” forbidden-food framing raises salience. He'd done this twice before, with the predictable result
Calorie counting the evening โš ๏ธ Addresses the wrong system and adds surveillance
โš ๏ธ "Healthy" substitutes on the sofa โš ๏ธ Would have kept the cue AND the eating. The point was to occupy the slot, not to relocate it
โš ๏ธ Anything at all at 21:40 โš ๏ธ The moment is the one place the intervention was guaranteed to fail

What happened

โš ๏ธ Week โš ๏ธ Evening episodes
Baseline โš ๏ธ 4โ€“5 / week, 11 years
1โ€“2 โš ๏ธ 2 / week โš ๏ธ Both on nights the tea didn't happen
3โ€“6 โš ๏ธ 0โ€“1 / week โš ๏ธ The tea had become automatic
โš ๏ธ 7 โš ๏ธ 4 โš ๏ธ A bad week at work; he bought the chips deliberately
8โ€“16 โš ๏ธ 0โ€“1 / week โš ๏ธ He did not treat week 7 as a failure. This is the important part
โš ๏ธ Month 6 โš ๏ธ Roughly monthly, and enjoyed โš ๏ธ Wanting fell as much as frequency did

โš ๏ธ The last row is the finding that surprised him.

"I assumed I'd spend the rest of my life wanting them and not having them. That's not what happened."

โš ๏ธ Cue-driven wanting is maintained by the cue. Removing the pairing degraded it. โš ๏ธ He had expected permanent deprivation and got extinction instead โ€” which is exactly what ยง33.8 predicts and is almost never how the trade is described to people.


โš ๏ธ Week 7, which is the case study's real content

โš ๏ธ Eleven years of previous attempts had all ended the same way: a bad week, a lapse, and abandonment.

โš ๏ธ What was different was ยง33.6 step 5.

โš ๏ธ Previously: lapse โ†’ "I have no willpower" โ†’ stricter rule โ†’ amplification โ†’ collapse.

โš ๏ธ This time: lapse โ†’ "the cue fired at a predictable moment during a stressful week, and I had bought the object" โ†’ no change to the rules โ†’ resumed.

โš ๏ธ He tightened nothing. โš ๏ธ That is the entire difference between an eleven-year pattern and a six-month one, and it is a change in INTERPRETATION, not in behaviour.

โš ๏ธ Which is why ยง33.12 argues that "this is not a character flaw" is a practical claim rather than a kind one.


Analysis

1. โš ๏ธ Four of five pieces of standard advice addressed the homeostatic system for a hedonic problem. โš ๏ธ They were not bad advice. They were the wrong tool (ยง33.3).

2. โš ๏ธ The cue was not the evening, the sofa, or the television. โš ๏ธ It was the end of a task, and it took a written log to find, because he had described it for eleven years as "the evenings."

3. โš ๏ธ The wanting/liking split was measurable and he had been treating it as evidence of irrationality. โš ๏ธ "I didn't even enjoy it" is accurate introspection.

4. โš ๏ธ Two behaviours that looked identical needed opposite interventions โ€” cue removal for the evening, a meal for the afternoon. โš ๏ธ This is the threshold doing practical work.

5. โš ๏ธ The intervention was made by unhungry Saturday Theo, in a shop. โš ๏ธ Not by 21:40 Theo, who had lost this argument approximately two thousand times.

6. โš ๏ธ Week 7's lapse was interpreted rather than punished, and nothing escalated. โš ๏ธ ยง33.6's step 5 is load-bearing, and this is what it looks like when it isn't loaded.

7. โš ๏ธ And the wanting decayed. The expected outcome was permanent deprivation; the actual outcome was extinction of a learned pairing.


Discussion Questions

  1. โš ๏ธ Four of five recommendations addressed the wrong system. Why is "drink a glass of water" so commonly given for cue-driven eating? What makes the homeostatic frame the default?

  2. โš ๏ธ "Ask yourself if you're really hungry" made things worse. Explain the mechanism. What should replace it? (ยง33.3.)

  3. โš ๏ธ He objected that removing the chips was "admitting I can't control myself." Is that objection wrong, or is it right in a way that doesn't matter?

  4. โš ๏ธ The tea does nothing nutritionally. Defend it as an intervention.

  5. โš ๏ธ Week 7 is the whole case study. What would have happened under the old interpretation? Map it onto ยง33.6's six steps.

  6. โš ๏ธ The wanting decayed rather than persisting. โš ๏ธ Why is deprivation the story people expect, and who benefits from that expectation?


Your Turn

Step 1 โ€” โš ๏ธ Log, don't judge. Two weeks. Four columns. No calories.

โš ๏ธ Time โš ๏ธ Context โš ๏ธ What started it โš ๏ธ Wanting 1โ€“10 / Liking 1โ€“10

Step 2 โ€” โš ๏ธ Find the seven-minute window.

โš ๏ธ Theo's episodes clustered within seven minutes of the same event. Look for a repeated TIME and then ask what happens just before it.

โš ๏ธ It is almost never the time itself. It is the transition.

Step 3 โ€” โš ๏ธ Split by signature.

โš ๏ธ Cue-initiated โš ๏ธ Interval-initiated
Wanting โš ๏ธ High Moderate
Liking โš ๏ธ Often LOW โš ๏ธ High
Time since food โš ๏ธ Short โš ๏ธ Long
โš ๏ธ Tool โš ๏ธ Change the situation โš ๏ธ Eat a proper earlier meal

Step 4 โ€” โš ๏ธ Occupy the slot.

The cue: __ โš ๏ธ What will follow it instead: __ (Small. Automatic. Not food-adjacent.)

Step 5 โ€” โš ๏ธ Make the decision in the shop, not the kitchen.

โš ๏ธ "The one thing I will decide about on a Saturday instead of at 21:40 is: __"

Step 6 โ€” โš ๏ธ And write your week-7 policy NOW, before you need it.

โš ๏ธ "When this fails โ€” and it will โ€” what I will conclude is: __"

โš ๏ธ If the honest answer is "that I have no willpower," you have just identified the step that ends every previous attempt. Write a different sentence and keep it.