Myths we retire · Science of Learning Institute

Retiring myths openly is part of what makes the rest of the site trusted. Each one below is graded Retired, with what to do instead.

Myth	What the evidence says	Do this instead
Evidence grade: Retired Students learn best in their "learning style" (visual, auditory, kinesthetic)	Properly designed tests of matching teaching to a style found no support; several contradicted it (Pashler et al. 2008(opens in a new tab))	Match the format to the content: maps for geography, sound for phonics
Evidence grade: Retired Working memory holds 7 ± 2 items	The real limit is about 3 to 5 chunks; Miller's "seven" was a rough estimate (Cowan 2001(opens in a new tab))	Plan for about four new items at a time
Evidence grade: Retired The learning pyramid ("we remember 10% of what we read, 90% of what we teach")	The percentages have no research behind them; every method on the pyramid can work in the right context (Lalley & Miller 2007, researchED briefing(opens in a new tab))	Choose methods by content and student knowledge, not by a retention chart
Evidence grade: Retired Rereading and highlighting are good study habits	A major review rated rereading and highlighting low; practice testing and spaced practice rated high (Dunlosky et al. 2013(opens in a new tab))	Quiz yourself; spread study out
Evidence grade: Retired Beginners learn best by discovering it themselves	Explicit teaching beat unassisted discovery across 164 studies (Alfieri et al. 2011(opens in a new tab))	Teach first, then apply (principle 2)
Evidence grade: Retired One effect-size ranking tells you what works	Rankings that average very different studies hide what was compared. Not used as evidence on this site	Go to the trials behind the claim

Sources