A psychologic screening battery, including assessment of perceptuomotor, attentional, memory, and emotional functioning, was presented to 134 patients at initiation of stroke rehabilitation. Significant differences in performance on tests based on side of involvement of stroke were noted. Poorer performance on sensory suppression errors and motor persistence score was found for individuals with left-side involvement and the opposite for word fluency score. Examining subjects with left- and right-side involvement separately using stepwise multiple regression, psychologic tests were able to account for 55% and 42%, respectively, of the variance in discharge outcome, as measured by the Barthel Index. Accounting for variance associated with change in Barthel Index score during rehabilitation proved more difficult, with R2 values of 0.14 and 0.07, respectively. Motor persistance score and half-hour recall proved to be the strongest of the psychologic tests in predicting rehabilitation outcome.