Put each user in a small weekly leaderboard ranked on points from normal use; top finishers move up a tier, bottom finishers move down. Duolingo reports big engagement gains — and its users report stress and unfair demotions.

+17% Overall learning time after Duolingo launched leagues, as reported by its former head of product; Duolingo's S-1 says “almost 20%”. Company-reported with the basis unstated, and not tied to a test.

3 cited sources, each with what it does not establish. Evidence from Duolingo.

Pick each reminder from a pool of written variants by how well each has worked, minus a penalty for ones this person saw recently. Duolingo's version beat random rotation; the penalty's own share is small.

+0.5% daily actives Duolingo's bandit against random rotation over a two-week online test; new users' next-day retention rose 2.2%. Group sizes not reported (Yancey and Settles, 2020).

3 cited sources, each with what it does not establish. Evidence from Duolingo.

Let a model predict how likely a learner is to recall each item, and put the items closest to being forgotten first — in review sessions and in the strength display. It helped recall in a classroom; at Duolingo it cut practice return.

+16.5% recall On an exam 28 days after the semester, over massed review (179 eighth graders, one Spanish course; points or relative not stated). At Duolingo, replacing the old Leitner system cut next-day practice return 7.3%.

5 cited sources, each with what it does not establish. Evidence from Duolingo, Lindsey et al., Tabibian et al..