The purpose of the current study was to investigate the effect of a peer mediation intervention on self-regulated learning of students with learning disabilities. The research captured both process and outcomes of peer tutoring following the tutors' experience of the intervention programme. Tutors' self-regulated learning was measured by modifications in mediation styles (process of tutoring) and capacity to benefit from adult mediation for Analogical thinking (outcomes) after participating in the intervention and peer tutoring exercise. The sample was composed of 162 pupils, 81 from grade 5 (tutors) and 81 from grade 2 (tutees or learners). Tutors were chosen from classes of children with LD as defined by the NJCLD; the tutees were randomly selected from regular classes. Each pair of tutors and tutees was randomly assigned to an experimental or control group. During the final tutoring session, the children's interactions were video taped and later assessed with the "Observation of mediation instrument". The tutors were also given a "Dynamic assessment analogies test". The research findings indicated that the experimental tutors showed higher self-regulated learning as compared with the control group, expressed by their mediation skills and improvement in Analogies scores between the pre-intervention and post-intervention phase. The findings are discussed with reference to Vygotsky's theory, self-regulated learning goals and today's heterogeneous classes. (Contains 2 tables and 4 figures.)