Considerable literature on the role of attentional focus in motor performance and learning has accumulated for over two decades. We report the results of comprehensive meta-analyses that address the impact of an external focus (EF, on intended movement effects) versus internal focus (IF, on movements of body parts) of attention on the performance and learning of motor skills. Values of effect sizes (ES) from 73 studies with 1,824 participants and 40 studies with 1,274 participants were used for examining the effects of EF versus IF on behavioral outcomes of motor performance and learning (separately for retention and transfer phases) respectively. The EF condition was more effective than the IF condition for performance, Hedges' g value = 0.264 (95% CI [0.217, 0.310]), retention learning, Hedges' g value = 0.583 (95% CI [0.425, 0.741]), and transfer learning, Hedges' g value = 0.584 (95% CI [0.325, 0.842]). Multivariable metaregression analyses on behavioral measures further indicated that neither age group, health status, or skill level, nor their two-way interactions, moderated the ES differences between EF and IF in performance, retention, and transfer models (all p > .100). A secondary analysis on 12 studies with 216 participants that examined the effects of EF versus IF on electromyographic outcomes of motor performance also indicated that EF was associated with more efficient neuromuscular processing, Hedges' g value = 0.833 (95% CI [0.453, 1.213]). From nine studies with 272 participants, performance measured by behavioral outcomes was found to be more effective when a more distal, rather than proximal, EF was used, Hedges' g value = 0.224 (95% CI [0.019, 0.429]). Overall, the meta-analytic results are consistent with prior narrative reviews and indicate that an external focus is superior to an internal focus whether considering tests of motor performance or learning, and regardless of age, health condition, and level of skill expertise. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

The complementary dominance hypothesis is a novel model of motor lateralization substantiated by decades of research examining interlimb differences in the control of upper extremity movements in neurotypical adults and hemisphere‐specific motor deficits in stroke survivors. In contrast to earlier ideas that attribute handedness to the specialization of one hemisphere, our model proposes complementary motor control specializations in each hemisphere. The dominant hemisphere mediates optimal control of limb dynamics as required for smooth and efficient movements, whereas the non‐dominant hemisphere mediates impedance control, important for countering unexpected mechanical conditions and achieving steady‐state limb positions. Importantly, this model proposes that each hemisphere contributes its specialization to both arms (though with greater influence from either arm's contralateral hemisphere) and thus predicts that lesions to one hemisphere should produce hemisphere‐specific motor deficits in not only the contralesional arm, but also the ipsilesional arm of stroke survivors – a powerful prediction now supported by a growing body of evidence. Such ipsilesional arm motor deficits vary with contralesional arm impairment, and thus individuals with little to no functional use of the contralesional arm experience both the greatest impairments in the ipsilesional arm, as well as the greatest reliance on it to serve as the main or sole manipulator for activities of daily living. Accordingly, we have proposed and tested a novel intervention that reduces hemisphere‐specific ipsilesional arm deficits and thereby improves functional independence in stroke survivors with severe contralesional impairment. image