Two hours every Sunday, or twenty minutes every day? Same weekly total, and it is not close: the daily version wins, and the reasons are among the most replicated findings in memory research. This is worth internalising, because the Sunday plan is the one busy people keep choosing, and quietly abandoning by November.
The memory case
The engine is the spacing effect, covered in depth in the spaced repetition article: material revisited across separate days consolidates far better than the same material massed into one sitting. Cepeda and colleagues' 2006 review of hundreds of comparisons found distributed practice winning across almost every condition tested. Daily study makes every day a review of yesterday: each session begins by retrieving slightly faded material, and that effortful retrieval (the testing effect) is precisely the workout that builds durable memory. The Sunday marathon gets one consolidation cycle a week and spends half its length in the diminishing returns of a tired brain. Attention research points the same way: the first twenty minutes are the best twenty minutes; the seventh twenty is mostly rereading with your eyes.
Sleep matters here too: memory consolidates across nights. Seven study days buy seven consolidation nights. One study day buys one.
The habit case
The second argument is quieter and possibly stronger: daily practice becomes automatic in a way weekly practice never does. Lally and colleagues' 2010 study tracking real habit formation found automaticity building over weeks and months of consistent repetition in a stable context; encouragingly, that missing a single day made little difference to the process. A daily anchor (“after breakfast, Italian”) stops costing willpower after a month or two. A weekly session stays a decision forever, and decisions get skipped. Language learning is a years-long project; the winning strategy is the one that removes the daily negotiation.
What twenty good minutes look like
Twenty minutes is enough to touch three of the four strands, if you spend them deliberately rather than opening an app and drifting:
- 2 minutes: fast review of known material, pure fluency work to warm up.
- 3 minutes: your error queue, the spaced items due today.
- 10 minutes: input: a chapter of a graded reader or a dialogue, meaning first.
- 5 minutes: output: three written sentences or a page read aloud.
I structured 
amo Italian around exactly this session (“twenty minutes a day” is literally the promise on the tin), with the lesson flow running warm-up, review, input and output in that order so the twenty minutes spend themselves. Not because twenty is magic, but because twenty daily minutes is the largest habit most working adults can actually keep for the years the project takes. And if you are wondering how many of those years: I wrote an honest answer in how long does it take to learn Italian.
References and attributions: Cepeda et al., Psychological Bulletin, 2006; Lally et al., European Journal of Social Psychology, 2010. Findings are paraphrased from the published literature with attribution. Part of the Science of Learning Italian series.
