Near-zero per-action cost makes loops hard to break

What it is

The individual actions composing a loop — a swipe, a tap, a pull — cost so little that each sits below any explicit decision threshold. This is the structural reason micro-action loops resist interruption: no single action is ever "worth" stopping to consider, so the decision to stop is never triggered.

Why it happens

Decisions to stop need salient triggers: boundaries, surprises, marked discomfort. Loops with near-zero unit cost suppress all three. Each step is too cheap to justify evaluating — running a cost-benefit analysis on a virtually free action is itself a loss. Outcomes are homogeneous and continuous, offering no surprise to break rhythm. Discomfort accumulates gradually and never crosses a "it hurts right now" threshold. The question "is the total investment worth it?" therefore never gets asked: it is an aggregate question, and the loop only ever presents unit prices. Five-minute "just one more" moments chain into an hour, but the hour never appears as an object of decision — what gets approved is always the next five minutes, and it is always approved.

Where it stops holding

Low unit cost does not mean low value: practicing an instrument or continuous reading are also cheap chained actions, and their loop-like quality is no problem — the criterion is correspondence between action and benefit, not the price of the action. The converse holds too: actions with real per-step investment naturally form decision points, and loops slacken on their own — which is why "add cost to each step" works as an intervention despite sacrificing efficiency. This entry concerns the unit price of actions within the operant loop; the granularity of the "continue" decision in scrolling feeds has its own dedicated treatment.

Applying it

To break a loop, raise the unit price back above the decision threshold: give high-frequency actions a real startup cost — choose a duration before opening, insert a confirmation after every N actions, or serve an unbounded supply in batches with an explicit restart action between them. To build beneficial loops, do the reverse: make the target behavior's gesture short and smooth. Verification: manipulate the unit price within the same loop (say, a two-second delay after every Nth action) and watch the loop-length distribution — a cliff-like drop from a tiny price increase is evidence the loop was being maintained by unit price.

Related

• Same group: P3.07.1 Anticipation itself evokes stronger neural responses than delivery · P3.07.2 Pull-to-refresh turns content updates into a lever action · P3.07.4 The criterion for addictiveness is behavioral-benefit decoupling
• Adjacent: P3.08.2 The decision cost of continuing to scroll is reduced to zero · P3.09.4 Friction must exceed impulse yet stay below genuine need
• Search terms: micro-action loop · action cost · commitment device