Interdisciplinarity and co-creation are becoming powerful catchwords in current debates about creativity and innovation in scientific knowledge production. Science policy initiatives and research funding agencies increasingly exhort scientists to undertake research approaches that are intersdisciplinary and integrate scientist and non-academic communities. However, despite its discursive prominence, we know little about the micro-mechanisms underlying knowledge production in academic practice that involves interdisciplinarity and co-creation, and to what extent these research practices contribute to creativity and innovation in science. Social network research has shown that social interaction is a powerful vehicle for knowledge creation and innovation (Granovetter, 1983; Hansen, 1999). This paper takes a relational approach to personal networks and focuses on how resources are mobilized through research networks (Borgatti and Cross, 2003; Levin and Walter, 2019; Walter et al., 2015). We investigate the specific benefits of research networks involving collaboration between actors belonging to different professional domains and institutional settings, and examine the specific type of resources mobilized by diverse research network configurations. This perspective contrasts with more standard network approaches that examine the structural aspects rather than the network mobilization decisions (Balkundi et al., 2007; Burt, 2004; Tan et al., 2015). Social network research shows that new knowledge accessed through network ties is crucial for the initial (idea generation) phase (Baer, 2012; Perry-Smith and Mannucci, 2017), but this is not the only benefit flowing through a tie. We draw on Levin et al. (2011) and examine several types of resources that can be mobilized through networks, including intangible (knowledge related and legitimacy) and tangible resources (materials and equipment, funding). Additionally, we take into consideration network diversity by examining two types of actor heterogeneity: cognitive (i.e., diverse professional practices) and institutional (i.e. diverse organization affiliation). By delving into the resources that flow through networks, we shed light on questions related to whether interactions with particular type of actors are conducive to idea generation and/or idea implementation processes; and about the specific type of benefits researchers gain when mobilize networks formed of highly heterogeneous actors. Answering these questions will provide valuable insights to understand the underlying collaborative dynamics in open innovation environments. Our theoretical claims are tested in the biomedical setting. This context provides a unique opportunity to examine the importance of personal research ties formed by academics, and how such ties provide diverse sets of benefits. Our study draws on a large-scale survey conducted in 2018 on Spanish biomedical scientists. Among other questions, respondents were asked to report their main professional contacts, along with the specific type of benefits(s) received from each contact. This allows us to depict our respondents¿ social structure of collaboration and to dig deeper onto the specific type of benefit received from each tie. The questionnaire was administered to 5,325 biomedical scientists. We received 1,146 valid responses, an overall response rate of 21.5 percent.