The Evidence on Teaching

Cognitive Training: A Field in Search of a Phenomenon

Gobet F, Sala G · 2022

grade Bcritiqueindependentnot-applicable
Sample
Response to Green et al. (2019), drawing on the accumulated meta-analytic record
Population
All ages
Design
The current state-of-the-field statement, and the most useful single source for understanding why the field keeps declaring the question open. Argues the persistent optimism results from neglecting the meta-analytic results and ignoring the statistical explanation - that apparent positive effects are sampling error plus artefacts.
Key findings
The overall far-transfer effect is null and there is little to no true variability between types of cognitive training - two claims that together remove the field's standard escape hatch ('it works for the right programme / the right population'). Recommends redirecting research to near transfer or to other methods of improving cognition.
Genetic confound
Low; the underlying evidence base is randomised.

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Far transfer, overallpooled meta-analytic effectnullstandardizedpost-trainingunclearend-of-treatmentfar-transfer
True variability between training typestau-squaredlittle to nonestandardizedpost-trainingnonenot-applicablefar-transfer

Cited by