Cue-Routine-Reward

Cue-routine-reward is the three-part structure of a habit, used as a diagnostic: name the trigger, name the behaviour, name the payoff — then change exactly one of them. The phrase comes from Charles Duhigg's The Power of Habit (2012).

Treat this page as the field procedure. The neuroscience behind why the three parts hold together — Graybiel's chunking work, and the finding that habits keep running after their reward has stopped being appealing — is on habit loop. Same three words, different job: that page explains the model, this one walks you through applying it to a behaviour you actually have.

Step 1: catch the routine in enough detail to be falsifiable

Most people can name their routine but not precisely. "I waste time online" is unusable. "I open a new tab and type the first two letters of a site name without deciding to" is a routine you can watch for.

Write it so specifically that an observer could tell you whether it happened. Vagueness here turns the next two steps into guesswork.

Step 2: isolate the cue with five questions

Duhigg's real contribution was less the model than the audit. Every cue falls into one of five buckets, so the moment the urge appears, record five things:

Question	Example answer
Where am I?	at my desk, second monitor on
What time is it?	15:20
What's my emotional state?	flat, slightly bored
Who else is around?	nobody, office half empty
What did I just do?	sent a long email

Do this for four or five occurrences and one column usually goes constant while the others scatter. That constant column is your cue. Honest caveat: this is structured self-observation, not a validated instrument. It tells you where to run your experiment, not what the answer is.

Wendy Wood and David Neal's work on context cues predicts what you'll usually find — location and the preceding action hold still far more often than mood does.

Step 3: test the reward, don't assume it

The obvious payoff is often not the operative one. The test is substitution: when the cue fires, do something else that delivers a candidate payoff, then check fifteen minutes later whether the urge is gone.

If a walk kills the urge, the payoff was a break from the screen. If only the original behaviour kills it, keep testing. One hypothesis at a time, a few days each.

What to change at each stage

Three named parts give you three levers with very different success rates.

• Change the cue — highest leverage, usually easiest. Remove it, move it, or move yourself. This is the lever Tomás pulls in the example on habit loop.
• Change the routine, keep cue and reward — Duhigg's "Golden Rule of Habit Change," and the only lever available when the cue is unavoidable. You cannot delete 15:20, or your own kitchen.
• Change the reward — hardest and least reliable, since you're arguing with something already learned. Mainly useful when building habits, where you can attach a payoff deliberately.

Example: Priya and the 15:20 vending machine

Priya buys crisps from the second-floor vending machine most weekday afternoons. She'd assumed hunger and tried bringing fruit for three weeks; the fruit went home uneaten and the crisps didn't stop.

Five days of the audit gave her this: location varied (twice from her desk, three times from the meeting room), mood varied (bored, stressed, fine), nobody was consistently around. Two columns held still — always between 15:10 and 15:30, and always after forty-plus minutes of uninterrupted screen work.

So she tested rewards. Day one: an apple at her desk. Urge back at 15:35, went to the machine. Day two: filled her water bottle at the ground-floor tap, two flights down, and walked back up. No crisps. Day three, same, no crisps.

The payoff was never food. It was standing up, moving, and four minutes off a screen after a long stretch of focus. Her fix kept the cue and the reward and swapped only the routine. Crisps went from roughly nine packets a fortnight to two — and both remaining ones fell on days a meeting ate the walk, which told her the meeting slot, not her willpower, was the next thing to fix.

How to apply this in Keel

The audit is the hard part, and Keel is where you keep it honest for the two weeks it takes.

1. Create the audit itself as a habit, not a resolution. Name it for the cue you're chasing — "Log the 15:20 urge" — and put it in the afternoon block so it appears in your day instead of living in your head.
2. Use completion-log notes as the five-column record. One line per occurrence: place, time, mood, who, what-just-happened. Five entries is usually enough for the constant to show.
3. Create the replacement routine as its own habit once you have a reward hypothesis, scheduled at the cue's time — "Water bottle walk, 15:15" — so the substitution is a scheduled action, not a resolve.
4. Read the misses as data. If the replacement fails only on Tuesdays, Tuesday has a structural problem. Bounce-back rate is the more useful number than a streak here, because a diagnostic phase is supposed to contain failed days.

If your cue turns out to be an existing reliable behaviour rather than a time or a mood, habit stacking is the better tool, and how to get back on track with habits covers what to do when a substitution collapses after a good fortnight.

Related terms

• Habit Loop — the concept and the neuroscience, rather than the procedure
• Implementation Intention — the if-then format for specifying a cue in advance
• Two-Minute Rule — shrinking the routine so the substitution is easy to start
• Habit Formation — how long a replaced routine takes to stick

References

1. Duhigg, C. (2012). The Power of Habit: Why We Do What We Do in Life and Business. Random House. Note: a journalist's synthesis of existing research, not primary work.
2. Wood, W. & Neal, D.T. (2007). "A New Look at Habits and the Habit-Goal Interface." Psychological Review, 114(4), 843-863.
3. Graybiel, A.M. (2008). "Habits, Rituals, and the Evaluative Brain." Annual Review of Neuroscience, 31, 359-387.