The spacing effect - discovered by Hermann Ebbinghaus in 1885 and confirmed in hundreds of studies since - shows that learning distributed across multiple short sessions produces dramatically better long-term retention than the same total time spent in one session. A 2006 meta-analysis in Psychological Bulletin (Cepeda et al.) found effect sizes around d = 0.5 to 1.0 - one of the largest, most reliable findings in cognitive psychology.
Habit formation research from University College London (Lally et al., European Journal of Social Psychology, 2010) found that simple daily behaviors take a median of 66 days to become automatic - with a range of 18 to 254. The key predictor of success is daily consistency in a stable context, not effort per session.
Game-based daily practice exploits two well-documented mechanisms: the dopaminergic reward of streaks (motivating return), and active retrieval (the act of recalling an answer strengthens the memory trace more than re-reading it would - see Roediger & Karpicke, Psychological Science 2006, the 'testing effect').
Practically, 5–10 minutes a day across multiple cognitive domains - memory, attention, logic, language - produces measurable improvements in those specific skills, and substantial gains in self-reported confidence and metacognition (Klingberg, Karolinska Institutet; Smith et al., JAGS 2009).