The Twenty-One Day Myth

What it is

Maltz was an American cosmetic surgeon, and Psycho-Cybernetics was his book about self-image. He observed that patients seemed to take about three weeks to adjust to their new appearance, and set this beside his observation that amputees experienced a phantom limb for about the same period. From such impressions he suggested that it takes a minimum of about 21 days for an old mental image to dissolve and a new one to form. Three things were lost as the idea travelled: it was a minimum, not a count; it described adjusting to a change, not building a routine; and it came from clinical impressions rather than a measured study. Writing on a UCL blog in 2012, Benjamin Gardner and Susanne Meisel suggested the myth may come from confusing habituation, getting used to something, with forming a new behaviour.

The loss can be watched inside a single book. Psycho-Cybernetics 365, a 2025 volume, carries the figure three times and cites nothing for it. Its January epigraph attributed to Maltz keeps the cautious form; the gloss on the same page turns it into a statement about how long mental change takes; by August it is a flat prediction that the habit will be in place. The surgical origin is never mentioned.

The measured evidence points elsewhere. In Phillippa Lally and colleagues' 2010 study of 96 volunteers, the time to reach a plateau of automaticity ranged from 18 to 254 days, with 66 usually quoted as the average. In 2023 Anastasia Buyalskaya and colleagues found that gym attendance took about four to seven months to become predictable and hospital handwashing a couple of weeks, and rejected the idea of a "magic number" of days.

In effect

Plan a new routine in months, not weeks, and do not abandon it when day 21 passes without it feeling automatic. A missed day made no material difference in Lally's data. Nir Eyal's Hooked puts it plainly: "research thus far has not found a universal timescale" (page 30).

What it does not say

It does not say a habit can never form in three weeks; the fastest participants in Lally's study reached their plateau in 18 days. The point is that there is no single figure, and the spread is the finding. Lally's measure is a feeling, and the curve fitted well for only 39 of the 96 starters. The page of the 1960 first edition carrying Maltz's sentence has not been settled, and the 1960 book was not read directly for this page; its wording comes from Gardner, Lally and Wardle's 2012 summary and other published quotations.

---