A Meta-analysis of the Executive Function components Inhibition, Shifting, and Attention in Intellectual Disabilities

This is a preprint of a manuscript that is currently under review. Analysis code and data can be found in the accompanying OSF project https://osf.io/njsd3/. BackgroundExecutive function is a concept for higher-order cognitive functions which havethe role of controller and modulator of cognitive abilities. The consensus inthe literature is that people with an intellectual disability perform significantlylower on executive function tasks than groups matched on chronological age.The comparison with groups matched on mental age is less clear. Therefore,the objective of this meta-analysis was to investigate to what extent executivefunction is impaired in people with intellectual disability compared to atypically developing control group matched on mental age. It was alsoinvestigated if the executive function component and intellectual disabilityaetiology moderated the effect.MethodsEligibility criteria were participants with intellectual disability (IQ 10. Working memory tasks and ratings of executive function were notincluded. The literature search yielded 6637 potentially interesting articles.26 studies (with 99 effect sizes) including 1395 participants were included inthe quantitative synthesis.ResultsA multilevel random-effects meta-analysis found that people with intellectualdisability performed statistically significantly lower than the mental age matchedgroup on the executive function tasks, g = -0.34, CI 95%= [-0.53,-0.16]. However, the heterogeneity between effect sizes was large. Theintellectual disability aetiology moderator was significant, but it only reducedthe heterogeneity marginally.ConclusionThe overall conclusion is that individuals with an intellectual disabilityhave more problems with executive function tasks than mental age-matchedcontrols. Limitations are the large unexplained variance and the remarkablyhigh number (69) of different tests that were used, which make more detailedconclusions problematic. This meta-analysis implies that future studies needto be of better quality, have higher power, and to a higher degree use thesame executive function tests.