Robot-mediated telerehabilitation has made great strides toward patient-tailored, cost-effective treatments. However, lack of motivation often leads to patients' non-compliance with their physical therapy regimen, undermining the potential advantages of telerehabilitation technologies. Embedding citizen science into robotics-based rehabilitation has been shown to increase patients' motivation to engage in otherwise mundane exercises. Here, we explore the feasibility of bolstering engagement in physical therapy through social interactions in citizen science context. We developed an online citizen science platform in which users work in pairs to classify images collected by an aquatic robot in a polluted water canal in Brooklyn, New York. The classification involves labeling objects that appear in the images and removing irrelevant labels. The application was interfaced by a haptic device for fine motor rehabilitation. We present preliminary results for the levels of engagement of healthy users performing the activity, with and without a cooperating peer.