Objective: To evaluate in a preliminary fashion whether several standard cognitive assessments predict treatment outcomes on real-world functional activities following Constraint-Induced Movement therapy (CI therapy) for either the upper extremity (UE) or the lower extremity (LE) for chronic stroke hemiparesis in the outpatient therapy clinic. Methods: 15 UE and 14 LE patients in the clinic underwent a short battery of cognitive assessments that evaluated sustained attention, episodic memory, executive control abilities, and general cognitive function. Spearman correlation analysis was used to evaluate whether each cognitive test predicted treatment outcome on the limb-specific Motor Activity Log (MAL). Results: Two assessments (delayed verbal memory and Trail Making Test form B) significantly correlated with LE MAL change that followed therapy. Conclusions: We tentatively conclude from this exploratory and preliminary study that cognitive performance may predict treatment changes in response to CI therapy for the LE. Moderate to large correlations that we observed between other cognitive assessments and CI therapy outcomes recommend replicating this study with a larger and more cognitively diverse sample of stroke patients to learn if these findings are generalizable. © 2006 - IOS Press and the authors. All rights reserved.