Objective: To evaluate the feasibility and effects of using motor imagery during therapeutic stretching in individuals with spasticity. Design: Randomized single-blind controlled pilot trial. Setting: Chronic day care unit, neurological rehabilitation unit and in the community. Subjects: Eleven individuals with spasticity in the arm requiring stretching as part of their normal routine. Interventions: In addition to their normal stretching routine, subjects in the experimental group received motor imagery during their stretches (n = 6). The control group received progressive muscle relaxation during their stretches (n = 5). The dose varied between 8 and 56 sessions over eight weeks. Main measures: Resistance to passive movement, measured with a torque transducer, passive range of movement, measured with an electro-goniometer, Modified Ashworth Scale (MAS) and level of discomfort during the MAS were assessed at baseline and after eight weeks by an independent assessor. These measures were recorded before and after a stretch intervention on both assessments. Results: Participants, therapists and carers tolerated the techniques well. Compliance was variable and adherence was good. Mixed ANOVA showed no difference over time and no difference between the motor imagery and progressive muscle relaxation group on the primary and secondary outcome measures (P>0.05). Conclusions: It is feasible to use motor imagery during therapeutic stretching. Statistical power was low due to the large variability in the population and the small sample size. Post-hoc sample size calculation suggests that future studies of this subject should include at least 54 participants per group. Further research is warranted.