Stroke survivors struggle to meet clinical recommendations for physical exercise duration and intensity. During the past two decades, music interventions have increasingly shown effectiveness in several motor tasks in stroke rehabilitation. Additionally, music has been found effective in increasing exercise performance in athletes and other clinical populations. Based on a meta-theoretical review paper by Clark, Baker & Taylor (2016), it was hypothesized that the therapeutic effects of music in physical exercise is modulated by the preference and task-specificity of the music. We will test this hypothesis using a three-armed randomized cross-over design comprised of the following auditory conditions: 1) a group-tailored playlist 2) radio as active control and 3) a non-music control condition during cycle ergometry cardiorespiratory exercise sessions. Participants are inpatient stroke survivors undergoing rehabilitation between 2-12 weeks post infarct.

Inclusion Criteria:

• between 2-12 weeks post ischemic or hemorrhagic stroke
• assessed eligible to participate in cardiorespiratory exercise by the patient's primary physiotherapist and interdisciplinary team -≥18 years old

Exclusion Criteria:

• uncontrolled medical contraindications for physical activity
• severe hearing deficit or cochlear implants
• severe deficit of both auditory and visual communication not allowing for guiding the patient in the exercise task