General knowledge is commonly described as a disposition rather than a body of learning. It is neither rare nor fixed. In psychometrics the relevant construct is crystallised intelligence — accumulated factual and verbal knowledge, distinguished by Cattell and Horn from fluid reasoning ability. The two follow different trajectories across the lifespan: fluid ability declines from early adulthood, while crystallised ability continues to rise into middle age and later, being a function of accumulated exposure rather than processing speed. General knowledge is therefore a domain that responds to deliberate study at any age, and the useful question is what form that study should take.
The domain is bounded and unevenly sampled
"Read more" is common advice and is difficult to act on, because it specifies neither a body of material nor a stopping point. The material asked about in quizzes is narrower than the material available. It concentrates in a stable set of categories — history, geography, science and the natural world, sport, music, film and television, food and drink, language — and within each, in recurring items: capitals and flags, monarchs and regnal dates, chemical symbols, Academy Award winners, number-one singles, football club nicknames. These recur in measurable proportions.
Two consequences follow. Revising evenly across everything knowable spends most of the available time on material that will not be asked. And revision weighted by observed frequency covers the asked distribution faster, for the same hours. The evidence for the method — that recall beats recognition and that spaced review beats massed review — is set out with citations in the guide to getting better at pub quizzes; it applies to general knowledge in full.
Testing supplies both practice and measurement
A regular quiz serves two distinct functions, and they are worth separating. The first is retrieval practice under realistic conditions, which strengthens retention directly (Roediger and Karpicke, Psychological Science, 2006). The second is measurement: an external, unflattering estimate of what is actually known, which self-testing at home tends not to provide, because the person setting the questions already knows which ones they can answer.
The output of the second function is the more valuable of the two. Questions answered incorrectly are, by construction, both within the distribution being asked and not yet known — the highest-value revision items obtainable. They should be recorded at the time, since recall of one's own errors decays rapidly.
Narrative structure aids retention
Isolated facts are harder to retain than facts embedded in a causal or narrative structure, which supplies additional retrieval routes. "The Berlin Wall fell in 1989" is a single cue. The sequence of events that brought it down supplies several, and the date is retained as a by-product. The same applies to the disappearance of the Aral Sea as an anchor for Central Asian geography, or to the year in which England had three kings as an anchor for 1066.
Background reading is therefore a complement to drilling rather than a substitute for it: drilling establishes retrievability, narrative establishes the connections that make retrieval robust. QuizMaxx publishes 364 short essays on this material, free and requiring no account.
What QuizMaxx does
QuizMaxx applies the above in one system. It supplies a curated body of general-knowledge questions ordered by measured frequency of occurrence in archives of real quiz questions, schedules reviews with an SM-2 spaced-repetition algorithm so that each item is revisited shortly before it would be forgotten, and requires typed answers rather than multiple-choice selection, so that practice is recall rather than recognition. It runs in the browser, needs no account and is free. Open QuizMaxx.
For readers already using flashcard software, the comparison with Anki sets out where each is the better tool.