PURPOSE: To determine whether a distributed mental practice (MP) format results in greater improvements in upper-extremity (UE) functional limitation and impairment compared with a traditional massed MP regimen.

RATIONALE: Stroke remains a leading cause of adult disability, with most of the expanding stroke survivor population retaining devastating UE motor deficits. MP constitutes a departure from other rehabilitative regimens in that actions are cognitively rehearsed in the absence of voluntary physical movements. It is widely known that the addition of MP to physical practice in hemiparetic stroke survivors significantly increases paretic UE use, kinematics, and function. However, it remains to be determined whether a distributed practice effect or spacing effect—which is well established to occur when physical practice is administered—is also exhibited during MP in stroke survivors.

DESIGN: Randomized controlled trial

PARTICIPANTS: Twenty-seven chronic stroke survivors

METHOD: The UE section of the Fugl–Meyer Assessment (FMA) and Action Research Arm Test (ARAT) were administered 1 wk before the start of the intervention. The rater was blinded and did not know to which group participants were assigned or if they even received an intervention. The intervention consisted of standardized physical practice and MP regimens, with one group administered massed MP (60 min of MP during a single daily session) and a second group administered distributed MP (20 min of MP occurring 3 times/day).

ANALYSIS: Unadjusted mixed-effects regression models, with repeated measures considered for time of assessment for each individual, were used to test the treatment’s effect on changes in FMA and ARAT scores at preintervention (PRE), postintervention (POST), and at the 3-mo postintervention visit (POST3). The Kenward-Roger adjustment to the degrees of freedom was used to control Type I error because of a relatively small number of individuals per treatment group. All p values and confidence intervals are presented at the nominal levels and are unadjusted for multiplicity.

RESULTS: After intervention (POST), both the adjusted FMA scores and the adjusted ARAT scores of participants in the distributed condition were significantly higher than those of participants in the massed condition (FMA: 95% confidence interval [CI] [0.82, 6.49], p = .01; ARAT: 95% CI [1.24, 6.67], p = .006). Three months after intervention, the distributed group showed significantly higher ARAT scores (95% CI [0.57, 9.22], p = .03); FMA scores were 2.96 points higher among the distributed condition, but not significant (95% CI [–2.31, 8.24], p = .26).

DISCUSSION: Results indicate that a distributed MP schedule is more efficacious in bringing about paretic UE changes than a massed practice schedule, especially in terms of reducing UE functional limitation.

IMPACT STATEMENT: MP is easily administered, requires no specialized equipment or therapist training, and can be used in many clinical or home locations, which constitute advantages over many traditional UE rehabilitative approaches. Our results not only further demonstrate the efficacy of MP in the poststroke population but also provide valuable insight regarding the exact method by which MP may be most effectively integrated into rehabilitative therapies.