5

Knowledge Base for Developing Effective Innovations

This session of the workshop brought in leadership, professional development, and team science research to identify key insights to incorporate into efforts and resources to support research leaders. Planning committee member Stephen Fiore, University of Central Florida, moderated the session and discussed relevant research on leadership and team performance. He then asked presenters Maritza Salazar Campo, University of California, Irvine, and Brian Uzzi, Northwestern University, to discuss research that can be used to shape approaches to training and professional development to advance research integrity, as well as critical issues that deserve more examination.

Fiore reflected that when scientists are promoted to positions as lab manager or director, they are running a small business while responsible for doing the science. As noted in comments throughout the workshop, these promotions often happen without training or other preparation. Interdisciplinary research adds another layer of complexity. Fiore pointed to an article published in Science on the increasing prominence of interdisciplinary research but the challenges of doing so within universities’ departmentalized framework and thus the need for patience and “a particular form of social intelligence.” Of note, the article was published in 1944 (Brozek and Keys, 1944). Fiore asked whether it is realistic to solve a problem identified almost 80 years ago. It is possible, he posited, thanks to dramatic changes in how policy, academic, and industry communities recognize the importance of collaboration and as research into the study

of teams has grown. Since the mid-20th century, several disciplines have emerged that study groups and teams, such as those embodied in the Society for Industrial and Organizational Psychology and the Academy of Management. Interdisciplinary organizations that focus on team work in science, such as the International Network for the Science of Team Science, have also emerged, and he urged looking to the experts who study the science of collaborations. As detailed in the National Academies study on team science (NRC, 2015), an evidence base is building on such issues as assembling teams, group dynamics in teams, supporting leadership development, virtual collaborations, study and measurement of science teams, promotion and tenure in science teams, and credit for team-based work.

He also pointed to multilevel perspectives when considering the science of team science and the responsible conduct of research (RCR). Structures at the micro, meso, and macro levels have different issues and perspectives that impinge on each other (Börner et al., 2010). The macro level refers to collaborations across such networks as centers, universities, and fields in which professional culture and identity, governmental policies, and societal values and norms must be managed. The meso level refers to the group or team, and issues to manage include group dynamics, interdependencies, and interpersonal skills. The micro level refers to individual scientists understanding what they need in terms of the science and, from an organizational perspective, what they need to know how to do (e.g., methods, procedures, and techniques).

UNDERSTANDING CONFLICT AND TRUST IN SCIENTIFIC TEAMS

Fiore commented on the frequent use of the term “teams” and offered a definition in the context of the workshop: Teams are characterized by multiple information sources and intensive communication, task-relevant knowledge with meaningful task interdependencies, affective and attitudinal factors that influence group dynamics, and coordination among members with specialized roles. Much of what is known about teams comes from researching them in other settings, such as the military. Fiore said understanding scientific collaboration as a form of teamwork to be both managed and mastered allows for leveraging this research (Fiore, 2008).

A critical dimension is to differentiate between team work and task work, he said (Fiore, 2008; Fiore et al., 2015). Task work refers to what needs to be accomplished to meet goals and complete objectives, or, as

Fiore said, the scientific work of the team. Team work, in contrast, refers to how they accomplish the task and refers to the attitudinal, behavioral, and cognitive factors to function as a team. Scientists have spent their entire careers learning and becoming better at task work, but it is important to help them become better at team work. Collaborative learning, professional development, and training on team work are being built into the education of scientists, but it is important to develop and measure interventions for RCR. Task-based and team-based training are both necessary. For example, in ethics, specific issues are the task-based content on which to train, while the team-based component might involve how to have difficult conversations when a violation occurs and how to create a psychologically safe environment.

Researchers in the organizational sciences have written about issues of conflict and trust, and their relevance to RCR. For example, one study (Hall et al., 2008) examined perceptions of interpersonal processes and found that trust was an essential prerequisite for effective collaboration in cross-disciplinary teams. Another study (Stokols et al., 2008) found that conflicts that arise from divergent disciplinary worldviews and perspectives can hinder the formulation of clear goals and their accomplishments. It is important to bring these issues to the fore, such as through the Toolbox Dialogue Initiative discussed in an earlier session (see Chapter 3). Awareness by team members of discrepancies, incompatible wishes, or irreconcilable desires can affect team work (De Dreu and Weingart, 2003).

Fiore differentiated between task-related conflict and team-based conflict. Task-based conflict should be cultivated, as it provides the intellectual give and take to hone ideas and build knowledge. Team-based interpersonal conflict, which is what most people think of when thinking about conflict, can lead to intolerance or problematic relationships that can harm RCR.

Trust is “the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor” (Mayer et al., 1995). Extending the discussion of tasks and teams, Fiore said both task and team trust are important for RCR (Fiore et al., 2015). To instill task-based intellectual trust, science team leaders must manage confidence in, or willingness to accept, and/or rely on each team member’s competence. To instill team-based interpersonal trust, leaders must manage confidence placed in members based upon feelings of security and levels of concern for each other.

Fiore raised two issues learned from the organizational sciences related to science team leadership. First, data show that leadership matters most

in teams that are highly interdependent. The higher the degree of interdependence, the more leadership can affect performance (Burke et al., 2006). Second, it is important to understand the nature of these science team interdependencies, which can be pooled, sequential, reciprocal, or intensive (Saavedra et al., 1993; Fiore, 2008). The former two are simpler, while the latter two are more complicated with shared outputs. In reciprocal and intensive interdependencies, people rely on each other in such a way that something is collectively produced that could not otherwise exist. Increasing collaboration size increases coordination challenges, he added.

Leaders can manage task and team issues related to conflict and trust in their teams, Fiore said. They can support task-related conflict by promoting usage of knowledge diversity and encouraging discussion of evidence. They can influence team-related conflict by creating a sense of psychological safety and managing emotional issues. They can support task-related trust by encouraging reliance on each other’s work-related judgments and dependence on each other’s task-related competencies. They can influence team-related trust by supporting the team to depend on each other for back-up in difficult situations and to confide in each other about issues affecting work.

INTERPERSONAL SKILLS TO LEAD TO CHANGE

Salazar Campo stressed the value to society when bright people come together to tackle social problems. Thus, drawing on her and colleagues’ work in the Team Scholarship Acceleration Lab at the University of California, Irvine, she urges shaking up the status quo related to the education and training of scientists.¹ Scientific progress will come “when we get the people side right,” she said. She related that she had recently participated in a panel for executives to talk about the impact of burnout and stress on productivity and effectiveness. She initially assumed that the audience would not be interested in so-called “squishy” topics related to the social sciences, but they were highly engaged. Because when people are motivated and driving toward outcome, efficiencies can be gained. A similar correlate can be made when scientists are equipped to lead, and manage, thus enhancing scientific productivity. Unfortunately, leadership development is often not a priority, and few scientists dedicate time and effort toward professional development.

___________________

¹ For more information, see https://tsal.uci.edu/.

To get people to follow and trust you as a leader in science, others must believe in one’s competence and expertise. Credibility is observed as an essential characteristic of leadership that others want to follow (Schein, 1996). In science, this means that followers are motivated to work with and for an investigator who is productive and at the forefront of their scientific fields. This scientific credibility is acquired through formal education and years of experience at the bench, in the field, or in community-engaged research. On the other hand, knowledge of how to lead is acquired more informally and tends to be shaped by scientists’ fields, advisors, and how they are socialized to lead and collaborate with others. When asked to describe exemplary scientific leaders, most will refer to those who trained them, and as scientists advance in their careers, they often draw on this reference as an important input into how they lead. This organic formation of leaders in science leads to loosely structured or evidence-based efforts to develop scientific leaders. In comparison, business leaders in industry are constantly asked to reflect on their values and paint a portrait of the leader that they want to be over the course of their career. Not surprisingly, most scientists are clear about what they want to accomplish in their field but lack a clear vision related to their leadership development goals and aspirations. Without intentional efforts to provide leaders with a toolkit and skills to manage and lead, few will form leadership vision that would enable them to catalyze action.

She added that answering a scientific question requires patience but, at the same time, takes place amidst strong performance pressures. Scientific leaders must often balance scientific curiosity while upholding ethical standards and leading their labs fairly and justly. Often investigators grapple with scarcity of resources, pressure to establish their own identities, the burden to take care of others in their lab, and to “win” in a competitive landscape. The result can be less attention toward upholding scientific standards or managing well. As such, the quest for novel findings, grant funding, and prestigious publications can undermine the motivation to dedicate time to leadership training that would support ethical management and capacity building of work units.

As a solution, Salazar Campo offered formalized training and development to address this competency gap. As shown in other fields, when well-designed systems of learning are created, organizational capacity is enhanced, and skills grow over time (Day, 2000). With the opportunity to see themselves as leaders, through the development of a clear leadership plan, scientists can understand blind spots, learn to communicate and lead better, and create individual trajectories for personal growth. She high-

lighted that development happens over the long term (Lant and Salazar, 2012), not through a one-hour workshop or other isolated activities. A multifaceted approach to growth, with ongoing training, coaching, journaling, and other methods, enables people to grow as leaders over time.

Salazar Campo urged a consideration of what is needed to upend the current ecosystem for science and to develop a sustainable intervention for this and future generations of scientists. She suggested that one possibility would involve multi-investigator training programs offered to entire labs, such that best practices could be passed from advisor to student, principal investigator (PI) to postdoctoral researcher (postdoc), and graduate student to undergraduate through joint training and modeling. As a consequence, training could potentially become irrelevant in the long term because best practices would become part of the culture. To reach this point, she offered two ideas: implementing evidence-based training and development programs and acknowledging and rewarding leadership and team impact. To make the first idea a reality, she proposed that federal grants require (or highly encourage) enrollment in a leadership or team science academy. Buy-in from key stakeholders is necessary for the scale and scope to build evidence that this type of training and development works. The second is to acknowledge and reward leadership and team impacts. If this type of leadership development is not incentivized in favor of individual accomplishments, it will not be part of the culture.

In sum, career enmeshment hinders leadership development in our institutions. Scientists often prioritize reaching specific career milestones or outperforming others, rather than focusing on personal growth and leadership capabilities, to the detriment of discovery. Training is needed to lead highly uncertain situations that differ from those in many other workplaces. When leaders are trained in forecasting skills, creative problem-solving, communication, strategic planning, and group processes, the outcomes can be profound (Dougherty and Hardy, 1996). Building these skills can result in a significant increase in interdisciplinary collaboration behaviors and communication and improve outcomes such as external funding and publication rates. Yet, to see these benefits, systems of incentivization and the availability of easily accessible training and development curriculum are necessary.

TEAM SCIENCE FROM A MACRO PERSPECTIVE

Uzzi described how big data can be analyzed to look at the science of team science and to understand how set-up strategies can give teams the

potential to be successful. He noted the value of the data but also the limitations in terms of missing many relationship and other skills needed to lead successful teams. He also spoke about gaps and concrete ways to put together a program of what should be considered as post-graduate education for scientists.

Since publication of Enhancing the Effectiveness of Team Science (NRC, 2015), it has become clear that teams dominate production of scientific knowledge, Uzzi said (Wuchty, Jones, and Uzzi, 2007). In addition, teams themselves are getting larger. He described a study of 26 million medical science teams that showed the modal size is six co-authors per paper. As mentioned by other presenters, he noted that teams benefit from diversity by discipline, but they also benefit from diversity of gender, age, skill, experience, and other characteristics. As an example, a paper that looked at same-gender and mixed-gender teams found that, controlling for many variables, mixed-gender teams performed better in terms of novelty of ideas and citation hits. However, looking at the gender of the lead and first authors, mixed-gender teams led by women did better than men in producing more novel results, but mixed-gender teams led by men have more citation hits (Yang et al., 2022). He suggested the role of diversity and leadership as an area of further study.

Networks are often developed through partnerships, and teams are embedded within networks. A multidisciplinary team may sound like a good thing, but he cautioned that if each team member has partnerships with the same third parties, the benefits of multidisciplinarity give way to more of an echo chamber. Networks provide a way to understand and measure this process and to understand leadership. The networks in which leaders situate themselves can help overcome their own limitations in certain skills or knowledge, which is borne out by the literature on networks and team science. Interdisciplinary teams require a “translator” who can communicate across disciplines, which is known as multivocality. He urged attention to the study of networks to understand science teams.

How people learn in teams is also critical, Uzzi continued. The structural approach that researchers have used when studying team science has focused on set-up strategies about what should be in place to make teams function. He noted recent investigations have looked at the role of mentorship, because scientists need the capacity to learn tacit knowledge from others (i.e., important information that is gained from “shoulder-to-shoulder collaboration” that cannot be explicitly taught). In one study, researchers looked at scientists who went on to win well-known prizes to see if they passed their

knowledge on to students. The study was set up to focus on students of the scholars before they won their prizes and compared student success to those who were protégées of non-prize mentors but who had similar records of productivity and impact. The research showed the probability of protégées of a non-prize-winning scholar have about a 2 percent chance to become a superstar and win prizes themselves. Students of prize-winning scholars have an 8 percent chance of being a superstar (Ma, Mukherjee, and Uzzi, 2020). This shows the value of the tacit knowledge imparted through the mentoring of prize-winning scholars even before they win their prizes, Uzzi said.

Uzzi recapitulated that teams dominate knowledge creation, diversity of teams drives team prosperity, networks help teams reach their potential, and mentorship is important. As he mentioned at the start of his presentation, the limitation of the literature is the emphasis on set-up strategies but less on social processes and soft skills. He commented that in his 20 years in the field, he has seen soft skills, once thought of as useless for business leaders, become something that students want to learn. With this demand, Uzzi said, there needs to be more research. There are also unknowns about the teams of tomorrow.

In teaching scientists to be better team leaders, he agreed with Fiore that most studies draw on research in other areas, such as the military or in sports. Uzzi’s model at Northwestern offers four levels of team effectiveness. Level 1 helps leaders understand they can only reach their potential through working with others. Level 2 focuses on team building and measurement. Level 3 focuses on systems with the goal to improve project management, fairness, and finances. Level 4 focuses on culture, with the goal to make the lab/team unique, which he suggested may be the most important feature of leadership as teams get larger. Looking ahead, Uzzi said these ideas will be most actionable when scientists understand that such training, which he described as post-graduate education not related to their specific area of science, is necessary for success. He suggested putting money into grants to allow lab leaders to participate in training and education to improve their skills. He also commented on the value of these programs to develop informal global connections and learning.

DISCUSSION

Fiore noted that there are many contemporary theories of leadership and asked Salazar Campo and Uzzi which theory might be most appropriate to enable RCR.

Salazar Campo commented that Uzzi’s structural macro approach and her more micro approach look at leadership through different lenses, and both are valuable. Rather than selecting a specific theory, she urged to understand the value of contexts and called for systems exchange, in which leaders are highly attuned to internal and external factors that impact their team. In terms of RCR, she noted the need to steward federal funding for research more responsibly, and a commitment to improving leadership skills means that funds are stewarded more responsibly. Uzzi commented that while all theories have a place, he and colleagues at Northwestern University do not use any specific theory. Instead, he said a new science of leadership focuses leaders on knowing how to plug into the world around them for resources, tips, and other benefits and how to use their unique competencies and skills. Leadership involves overcoming limitations to become more innovative and effective. The common denominator is how to build better networks and to connect between networks and achievements. Salazar Campo posited that while scientific leadership is about filling deficits through networks or teams, it is also about inspiring action to solve complex problems beyond complementarity.

When she works with department chairs and center leaders to conduct individualized coaching, she finds they are sometimes struggling between operation of the lab and the strategic future of their science. They could not find time for creativity amid their overbooked calendars. In these cases, she helps them clear the space to spark new ideas so that they can continue to lead.

REFERENCES