Within the field of developmental science, there is a general agreement of the need to work together across academic disciplinary boundaries in order to advance the understandings of how to optimize child development and learning. However, experience also shows that such collaborations may be challenging. This paper reports on the experiences of bidirectional collaboration between researchers in a multidisciplinary research team and between researchers and stakeholders, in the first randomized controlled trial in Swedish preschool. The objective of the trial was to investigate the effects of two pedagogical learning strategies evaluating language, communication, attention, executive functions and early math. The interdisciplinary team includes researchers from early childhood education, linguistics, developmental psychology and cognitive neuro science. Educational researchers and theorists within the field of early childhood education in Sweden have during the last two decades mainly undertaken small-scale qualitative praxis-oriented and participative research. There is a widespread skepticism with regards to some of the core principles in controlled intervention methodologies, including a strong resistance towards individual testing of children. Consequently unanticipated disagreements and conflicts arose within the research team, as RCT methodology requires the measurement of effects pre and post the intervention. The aim of this article is to discuss the conditions for bidirectional collaboration both between researchers and stakeholders and between researchers in the research team. The findings illustrate strategies and negotiations that emerged in order to address ontological and epistemological controversies and disagreements. These include (a) the negotiation of research ethics, (b) making divergences visible and learning from each other, (c) using a multiepistemological and methodological approach as a complement to the RCT design and (d) the negotiation of research problems that are shared between educators and researchers. Background and aim For more than 20 years, educational researchers and theorists in Sweden have pushed for a group-based investigative learning approach, which takes its starting point in children's sense of wonder in their encounter with the world and intrinsic motivation to learn (e.g., Dahlberg,