Does Far Transfer Exist? Negative Evidence From Chess, Music, and Working Memory Training

Three meta-analyses: chess instruction, music instruction, and working-memory training

Children

Tests the same hypothesis three times through independent literatures, which is why it belongs in a debunked-layer file rather than only in a music or chess topic: the design-quality gradient reproduces identically in all three.

Small to moderate raw effects on cognitive and academic skills, but effect sizes are INVERSELY related to the quality of the experimental design, in particular the presence of an active control group. The authors conclude far transfer of learning rarely occurs and extend the point to spatial training, brain training and video games.

Medium at the level of the raw estimates - children selected into chess and music differ - and this is part of why the design gradient matters. Low for the active-control estimates.