The objective of this project is to understand the role of awareness in the initiation and sustenance of research-oriented virtual organizations. Research-oriented virtual organizations enable: 1) interaction between researchers with diverse perspectives who might not otherwise work together; 2) the sharing of expensive and scarce resources; and 3) novel ways of accomplishing tasks and solving problems. Despite these advantages, one frequent problem encountered by members of research-oriented virtual organizations is persistent difficulty in communicating and coordinating with remote colleagues. One key reason for this is that many research-oriented virtual organizations lack the capacity to support effective awareness and interaction among their members as projects progress.
This work builds on theories of social networks and transactive memory as well as theories of interpersonal awareness in focusing on what we call ?awareness networks? ? that is, networks of individuals who, at some level of detail, keep track of each other. The work focuses on 3 key issues: 1) developing a novel two-staged approach to awareness that bridges the current theoretical and practical gap between the initiation of research-oriented virtual organizations and sustained effective work; 2) developing an understanding of the roles of network ties in interpersonal awareness that allow us to apply a network- and relationship-based understanding of individuals to the larger problem of designing tools to provide interpersonal awareness and expertise information in research-oriented virtual organizations and communities; 3) extending transactive memory theory to include multiple dimensions of awareness beyond simple knowledge of expertise distribution.
This work will result in new theories and technologies that inform and enhance our ability to support virtual organizations that address key challenges confronting society. Prospective and current ROVO members will be better able to locate each other, coordinate and collaborate toward successful joint outcomes.
Increasingly, solutions to important social, research and policy problems are achieved only when individuals with diverse expertise from multiple geographic locations are able to collaborate effectively. The goal of this project was to better understand how "virtual organizations" – that is, collections of people working together across organizational and geographic boundaries – of researchers are formed and how they can be sustained. By studying real-world virtual organizations focused on medical research within Cornell University, we aimed to enable higher rates of success in future virtual organizations and to spearhead the development of new technologies to support collaborations such as these. What We Did Our work had three primary components. Field Study of Existing Virtual Organizations We spoke with and observed 31 researchers involved in geographically distributed collaborations between Cornell University’s medical school in New York City and the primary Cornell campus in Ithaca, NY. Data from these interviews and observations were transcribed and analyzed by a team of researchers, yielding valuable insights (see below) on how these collaborations formed and what factors led to their ultimate success or failure, as perceived by the participants. Quantitative Analysis of Virtual Organization Outputs To better understand and describe the full scope of distributed collaborations at Cornell – and compare their frequency and outputs with local collaborations on a single campus – we quantitatively examined data about the publication history of 3543 Cornell faculty members. We used these publication records to compare publications that had authors on one campus to publications that had authors on multiple campuses. We incorporated data about collaboration catalysts, such as seed grants and topic-focused retreats, to assess their effectiveness. Developing a Technology Testbed Based on findings from this project and related work, we experimented with the development of next-generation awareness and interaction technologies. Using the OpenMessenger framework developed by one of the PIs, students explored ways this framework could be adapted for use in the contexts described by our participants. Initial experiments involved using an eye-tracking device for a peripheral awareness display, and using the Microsoft Kinect sensor for awareness purposes. We are excited by the possibilities presented by these systems, and have a working prototype system. Intellectual Contributions Our key intellectual contributions are: In our field study, we found first that participants were generally enthusiastic about collaborating, but extremely apprehensive about starting new collaborative projects. They were reluctant to engage with new collaborators unless they felt that the experience was likely to be a positive one. Indeed, the immediate relevance of topics and expertise often took a back seat to compatibility in work styles and approaches. They tended to be most interested in working with people with whom they had shared past experience, but also sought to assess whether new potential collaborators were likely to be good or not. We also found that – even within Cornell University – differences in institutional policies and culture across the two campuses often served as an impediment to collaboration, a factor that would likely be amplified in collaborations across institutional lines. In our analysis of publication records, it was clear that there is substantial interest in distributed collaboration, as it is occurring with greater frequency; and that institutional efforts to facilitate distributed collaboration seem to be having some effect. One thing to keep in mind, however, is that the distribution of these efforts is not at all even across departments, or even across faculty within departments. There is clear evidence that relatively small numbers of departments, and faculty within those departments, are accounting for a substantial fraction of the activity. Institutional efforts, such as retreats and seed funding, seemed to be most effective for those individuals who are highly central in the cross-campus co-authorship network. Broader Impact Our work has broader impact in two key respects. First, the results directly inform efforts to encourage collaboration in addressing key research problems in society today. We have presented these results to researchers and administrators involved in these situations, and have published our results in relevant conferences and publications to make them further available. Second, our work involved the training of two doctoral students, one postdoctoral researcher, and several undergraduate students in STEM fields. These researchers gained valuable skills in conducting research, analyzing data and developing novel technologies. Students involved in this project have gone on to pursue graduate study in related fields, work in the technology industry (e.g., Google), and in tenure-track academic jobs.
