Case Study 1 — Jordan: The 3 AM Problem Has a Name

The Audit

Jordan had been tracking his sleep since reading the chapter — not obsessively, but with the same empirical curiosity he'd been bringing to everything else. Two weeks of a simple log: bedtime, estimated time to sleep, middle-of-night wakings, wake time, quality rating (1–5), how he felt at 8 AM (1–5).

The data was instructive.

Average total sleep: 6 hours 22 minutes.

Variance: high. Wednesday nights, before early Thursday meetings, averaged under 6 hours. Weekend nights averaged over 7. The Sunday-to-Monday shift — sleeping later Saturday, then trying to sleep at a workday time Sunday night — reliably produced poor Sunday sleep and a difficult Monday morning.

The 3 AM wakings appeared most often: on nights with less than 6 hours before the waking, when he knew he had an important meeting the next day, and on nights when he had done substantial work email after 9 PM.

He looked at the correlations. The email pattern was clear: nights when he checked work email after 9 PM showed a 3 AM waking rate of roughly 60%. Nights when he didn't: roughly 20%.

He showed Dev the log.

Dev said: "You've been doing this for fifteen years and just now looked at the numbers?"

Jordan: "I didn't have the tracking protocol."

Dev: "You're a data analyst."

Jordan: "Personal data is different."

The Architecture Problem

Jordan understood sleep staging from the chapter, and the explanation for the 3 AM pattern clicked immediately.

He was a morning type — he had always known this, but had framed it as a preference rather than a biology. He naturally woke early (6–6:30 AM even without an alarm), was cognitively sharpest before noon, and experienced his clearest planning thinking between 7:30 and 10:30 AM. After 2 PM, his ability to do genuinely original analytical work declined noticeably.

What he hadn't understood: his early chronotype meant that his REM-rich late-sleep hours — the 5-7 AM window for most adults — occurred earlier in the night, roughly 4–6 AM. When he woke at 3 AM and lay awake for an hour, he was often disrupting the beginning of his primary REM window.

And the REM-deprived version of Jordan was, as the chapter describes, more emotionally reactive, less creative, and worse at the associative thinking that strategic work requires. The morning meetings that felt most productive to him weren't just a preference — they were biologically timed.

He had been scheduling 8 AM meetings because he was "a morning person." He had not been protecting the 4-7 AM sleep window that made the morning meetings possible.

The Changes

Jordan made a set of specific changes, designed iteratively over six weeks.

Change 1 — The 9:30 PM email cutoff (firm)

He had already tried versions of this. What was different now: he changed it from a personal rule to a structural constraint. The work email app on his phone was moved off the home screen, notifications disabled after 9:30, and he added a phone filter that blocked work email push notifications between 9:30 PM and 6:30 AM.

This required a conversation with Rivera about availability expectations. Rivera was, predictably, fine with it. "I wasn't expecting you to respond at 11 PM," Rivera said. "Were people actually expecting that?"

Jordan didn't answer that question directly.

Change 2 — Consistent sleep and wake time (including weekends)

The social jetlag pattern was the most surprising finding from the log. He hadn't thought of himself as having weekend sleep drift — but the data showed a 70-minute average difference between weekday wake time (6:15 AM) and weekend wake time (7:25 AM). Enough to produce meaningful circadian misalignment by Monday.

He negotiated with Dev: weekend wake times within 45 minutes of weekday times. He explained the social jetlag concept. Dev, who had always slept more consistently, looked slightly smug.

Change 3 — Caffeine cutoff at 1 PM

This was harder than expected. Jordan's second coffee was typically at 3 PM — a habit built around his afternoon energy trough. The 3 PM coffee was now revealed as partly masking a genuine alertness trough with stimulant chemistry that was still active at 9 PM.

He replaced the 3 PM coffee with a ten-minute walk. This was not as satisfying. It was, measurably, better for his evening.

Change 4 — The 3 AM recovery protocol

For the inevitable nights when he did wake at 3 AM, he designed a response protocol rather than relying on the previous approach (lying in bed, rehearsing the next day's obligations, and spiraling gradually deeper into the 3 AM accounting session).

The protocol: if not asleep within 20 minutes, get up. Go to the kitchen. Make a small cup of herbal tea (no caffeine). Sit at the kitchen table without a phone or screen. Read from the physical book currently on the table (he started keeping a book there specifically for this purpose). Return to bed when genuinely sleepy.

The protocol came from the CBT-I principle of stimulus control: the bed should be associated with sleep, not with lying awake. By getting up and returning when sleepy, he was gradually retraining the association.

The Six-Week Results

Jordan's sleep log at six weeks showed:

Average total sleep: 7 hours 1 minute (up from 6:22). 3 AM waking frequency: approximately 35% of nights (down from approximately 50%). Sleep quality rating: 3.6 average (up from 3.1). 8 AM energy rating: 3.8 average (up from 3.3).

Not a transformation. A measurable improvement.

The change he noticed most was not the numbers — it was the quality of his first two hours of the working day. The 7:30-9:30 AM window, which had always been his cognitive peak, felt sharper. He was doing better analytical work in that window than he had been doing. He was also more emotionally regulated in his first interactions of the day — less reactive to the small irritations that had previously set the tone for the morning.

He made one scheduling change at work: he moved his most demanding analytical work — the work that required original thinking, not review or meeting — to the 8:30-10 AM block, protected from meetings. Anything that was meeting-compatible could go to the afternoon.

Rivera noticed within two weeks: "Your comments in Thursday's sessions have been different lately. More specific."

Jordan: "I've been protecting the morning."

Rivera: "Whatever you're doing, keep doing it."

The Children Conversation (Again)

In the seventh week after the sleep changes, Jordan and Dev had a longer version of the children conversation — not a decision, but a genuine exploration.

Dev had been working through their own version of the same questions. The independent work pivot (now disclosed, still in planning) had clarified something: Dev wanted to have more time and control over their schedule. A child would affect that timeline. Dev wanted to know whether wanting a child was still true if the conditions changed.

The answer, for Dev, was yes.

Jordan's answer was harder.

What he noticed, in the conversation, was that he could actually track his own emotional response more clearly than before. The previous versions of this conversation had involved a kind of fog — he knew he was uncomfortable but couldn't locate the specific discomfort. In this conversation, he could feel the specific quality of the fear more precisely: not fear of children, not fear of parenthood in the abstract, but fear of the loss of control that parenthood represented. The 3 AM accounting sessions — which had reduced but not disappeared — were, he now understood, partly his nervous system's way of maintaining the illusion of control over the variables in his life. A child was a variable he could not control.

He said this to Dev, more precisely than he had said anything about it before.

Dev listened. Then: "I think that's the most honest thing you've said about this."

Jordan: "I've been sleeping better. My emotional vocabulary apparently improved."

Dev: "That's either a joke or a very interesting observation."

It was both.

Discussion Questions

1. Jordan's sleep log revealed that the 3 AM waking pattern was correlated with evening email checking and high-variance weekend sleep timing. What does this illustrate about the relationship between behavior patterns and sleep quality — and about the value of data in understanding problems you've lived with for years?

2. Jordan made structural changes (app settings, negotiated availability norms) rather than relying on personal resolve. How does this connect to the environment design principles from Chapter 29?

3. The CBT-I stimulus control principle — get up if not asleep within 20 minutes, return when sleepy — is counterintuitive for most people who believe staying in bed is the right response to insomnia. What is the mechanism by which the counterintuitive approach works?

4. Jordan noticed that better sleep improved the precision of his emotional self-awareness during the children conversation. The chapter describes the prefrontal-amygdala decoupling produced by sleep deprivation. How does improved sleep quality change the access to fine-grained emotional experience?

5. Rivera noticed Jordan's changed quality in Thursday strategy sessions without knowing about the sleep changes. What does this suggest about the relationship between sleep and the kind of thinking that is most visible in professional contexts?