Transcript
Transcript: Reshaping Scientific Networks Through Collaboration
[00:00:01 Text appears on screen: "Welcome."]
[00:00:06 The screen fades to Cheryl Engler Wadasinghe.]
Cheryl Engler Wadasinghe (Executive Faculty Member, Canada School of Public Service): Good afternoon, everyone, and welcome to our session. My name is Cheryl Wadasinghe and I'm an executive faculty member at the Canada School of the Public Service. I will be your session moderator for this presentation today, which is entitled Reshaping Scientific Networks Through Collaboration.
Before we begin, I would like to recognize that I'm speaking to you from the traditional, unceded territory of the Algonquin Anishinaabe people. I want to express my gratitude to the generations of Algonquin people past and present as the original caretakers of the space I occupy. I was fortunate recently to learn about the teachings of the seven grandfathers, which hold great significance to the Anishinaabe people. One of the seven grandfathers is wisdom. The others are love, respect, bravery, honesty, humility, and truth. It is said by the teaching on wisdom that to cherish knowledge is to know wisdom. This teaching resonates with today's session which underscores the importance of bringing different kinds of knowledge together to be used for the good of all people. I recognize that participants have joined this session from various parts of the country and therefore work on a different Indigenous territory.
We'll start today with a presentation by our speaker, followed by a question and answer session. I want to let you know that you can submit your questions now and throughout the session by clicking the chat function icon located at the top right corner of your screen. You can submit your question in the official language of your choice and we will get to as many of them as we can. Tackling today's complex scientific and societal challenges requires researchers to collaborate across disciplines, institutions, and geographic locations. Understanding how these collaborations form and grow is key to creating space for innovation and supporting effective research and policy outcomes. Today's session will explore how government funding initiatives are reshaping collaboration among researchers.
[00:02:09 Shiri Breznitz appears in a separate video chat panel.]
Drawing on her research, our speaker, Dr. Shiri Breznitz, will highlight how scientists are coming together from areas such as biology, engineering, computer science, and ethics to advance scientific discovery and innovation. It is my pleasure to introduce Dr. Shiri Breznitz. Dr. Breznitz is the Roz and Ralph Halbert Professor of Innovation at the Munk School of Global Affairs and Public Policy and is cross-appointed to the Strategy Area at the Rotman School of Management at the University of Toronto. She also serves as the Director of Research at the Munk School. Dr. Breznitz is the 2026 Jocelyne Bourgon Visiting Scholar at the Canada School of the Public Service and is a globally-recognized expert in economic geography, specializing in innovation policy, technology commercialization, and regional economic development. Her research explores how innovation systems evolve and how public policies related to universities, intellectual property, and research networks can support inclusive and resilient economic growth. With that said, please join me in welcoming Dr. Shiri Breznitz.
Dr. Breznitz, the floor is yours.
Shiri Breznitz (Visiting Scholar, Canada School of Public Service): Thank you Cheryl. It's very nice to be here, virtually but here with all of you, and I'm happy to share with you this paper today which is, as Cheryl mentioned, looking at specifically how policy impacts social network between researchers.
[00:03:33 A slide is shown with the text:
"Munk School of Global Affairs & Public Policy Innovation Policy Lab – University of Toronto"
"Reshaping Scientific Networks Through Collaboration"
"Shiri M. Breznitz, Roz & Ralph Halbert professor of innovation, University of Toronto"
"***In collaboration with Paige Clayton & Mircea Gherghina".]
We're going to look at different kinds of collaborations, but I think for your interest, the important part is the impact of policy, and so… it's not moving. Hold on. Okay, we made it.
[00:03:52 A slide is shown with the text:
"Highlights:
We explore whether and how collaboration networks of a sample of principal investigators (PIs) changed during a specific funding initiative.
Our research setting and sample come from a group of researchers who participated in a government funding program that sought to incentivize interdisciplinary research in a relatively novel field of science: regenerative medicine.
Our findings demonstrate how targeted funding initiatives transform collaboration networks based on their individual, organizational, and geographical characteristics, fostering interdisciplinary connections and driving research in emerging fields."]
Okay. So, today, we are specifically looking at these collaborations between principal investigators and how they changed during a specific time of funding, and we're looking at researchers that are working in a relatively, or when they started working was relatively, a new field, regenerative medicine, and I can already tell you, before we go through the entire presentation, that we find in this work that the funding initiatives transform the collaborations of these researchers, both on their individual characteristics but also organizational-wise and geographical-wise, which I think is really interesting when we try to think about how policy can support research.
[00:04:52 A slide is shown with the text:
"Collaboration in Research:
Kinds of Collaboration – Co-authorship of publications (Li et al., 2013), Patent co-investorship (Balconi, Breschi & Lissoni, 2004), Joint projects with industry partners (Melese et al., 2009), Research grants (Smith, Sarabi & Christopoulos, 2023)
The number, diversity, and intensity of ties results in higher research productivity and impact (Gonzalez-Brambila, Veloso & Krackhardt).
Tie formation depends not only on individual initiative but also on broader structural conditions, such as the emergence of new scientific fields (Citron & Way, 2018)."]
So, before we go into the results of this study specifically, I want to tell you a little bit about collaboration in research, okay? So, there are different kinds of collaboration. The most known is, of course, collaboration in publications. We tend to co-author papers with other people, with our students, with colleagues. We also apply for grants together. We patent together. So, sometimes when you look at patents, there are several inventors listed. And of course, we do projects with industry. Now, what we know from research is that the number of people you collaborate with, how many times you collaborate with, so you can have one collaboration with person X but five papers with person Y, the more you have of these collaborations, the more intense they are, the higher research productivity you're going to have and its impact is also going to be higher, okay? Now, what we also know on different aspects of these studies on collaboration in research is that these tie formations are not dependent only on the individual itself, himself, him or her. It's also about the broader conditions, right? So, how does the organization which I am working in treat collaboration, right? The entire environment is really important, but it's also really unique when we start to think about a new research field. Why am I saying that?
[00:06:37 A slide is shown with the text:
"The Development of a New Field:
Knowledge is at its highest level of uncertainty, and researchers must pool diverse expertise and resources in order to create breakthrough science (Camuffo et al., 2020; Schumpeter, 1939).
Building such collaborations often requires searching beyond immediate networks to span different organizations and geographies, a process that entails both direct costs (e.g., attending conferences, networking) and opportunity costs (e.g., foregone research in existing areas).
Government funding can help offset these costs by providing resources and flexibility to build new connections (Hicks et. al., 2019) – Results are not always positive."]
Because when you look at knowledge, okay? When we are looking at new breakthroughs, new knowledge, this is the time where you need the highest level of diverse expertise. So, if you think about it, if we are working with the same people all the time, we are working with the same pool of knowledge constantly. So, when we are developing a new field, it's really important to get some new information in, and building these kinds of collaboration needs more than the traditional organization and geographies that, for example, I'm used to working with, okay? So, if I constantly work with the same co-authors or universities, I probably will not start something very revolutionary in my work, okay? Now, what we also know from research is that government funding can offset some of these risks and costs that are related to starting new projects and new connections, and this is part of what we are interested in this project.
[00:07:45 A slide is shown with the text:
"Geography & Ties:
Scientific collaborations follow certain geographic patterns given where new science has been well resourced (Isfandyari-Moghaddam et. al., 2023).
Agrawal et. al. (2015) show that the average physical distance between collaborators has increased over time, suggesting a growing dispersion of research networks.
Lungeanu and Contractor (2015) observe strong country homophily even in an emerging interdisciplinary field."]
So, what do we know about geography and social networks, okay? If we look at scientific collaborations, we see that they follow certain geographical patterns, okay? So, for example, I did my PhD in Cambridge which means I have connections in Cambridge. I worked in the U.S., I have connections in the U.S. So, you can follow my career to see who are the people I am collaborating with, okay? We also know that people kept saying, once we have the internet, geography won't matter. So, yes, we can see that doing presentations or meeting on Zoom helps us reduce the need for being in the same office or in the same lab. However, we do still find that geography matters, right? And it matters a lot when you think about the same culture, right? So, people who are coming who have the same culture, and that doesn't mean just being in the same country but academic culture, thinking about the kind of research you're doing, so working with people who have similar culture matters.
[00:09:06 A slide is shown with the text:
"Organizational Foci & Ties:
Organizational co-location with other startups in a business incubator increased network size compared to startups that participated as members in the incubator but did not become tenants (Breznitz et. al., 2018).
Co-working spaces for entrepreneurs provide connections and social support (Howell, 2022)."]
The same thing applies when we think about the organization in which people operate in, okay? So, co-locating in the same organization with other people matters. So, in this study we've done at Georgia Tech looking at an accelerator there, we found that startups that were locating in the same business incubators actually increased their networks in comparison to startups that were supposedly part of the same network but did not locate in the same accelerator, okay? The same thing happens with co-working spaces. Being with people in the same location, if you think about what I just said before, even with Zoom, even with all of the digital tools we have, it still matters to be close to people.
[00:09:53 A slide is shown with the text:
"PI Characteristics – Gender:
Female researchers, especially those in non-tenure track roles, collaborate more with other women compared to men at similar ranks (Bozeman and Corley, 2004).
Women engage more often in interdisciplinary research (Van Rijnsoever and Hessels, 2011).
Women participate less in international collaborations (Uhly et. al., 2015).
Women engage in more cross-disciplinary collaborations outside their home disciplines (Hall et. al., 2018).
Women's ties may remain denser and more repetitive than men's (Ductor et. al., 2023)."]
Now, what about the specific individual characteristics? So, I'm going to go through a few things here that we looked at in the research. So, the first thing is gender, okay? So, we know there's differences between female and male when it comes to entrepreneurship but it's not much different when you think about it in social networks in research, okay? So, female researchers, especially if they're not tenured, right? Tend to collaborate more with other women, rather with men. They tend to also be engaged with more interdisciplinary research. They less collaborate with international collaborators. However, they collaborate more often cross-disciplinary outside their home disciplines, so much more likely to find women doing these multidisciplinary networks. And of course, their ties tend to stay much denser and repetitive, so they tend to work with the same people over time.
[00:11:05 A slide is shown with the text:
"PI Characteristics – Professional Status:
Star scientists attract a disproportionately larger pool of collaborators (Agrawal et. al., 2013; Azoulay et. al., 2010).
Assistant professors collaborated more than other ranks within their university, while higher ranked full professors tended to have greater international collaborators (Abramo, D'Angelo & Murgia, 2014).
Full professors develop stronger connections to industry in particular, due to their longevity and status, compared to others (Abramo, D'Angelo & Murgia, 2014, Van Rijnsoever, Hessels & Vandeberg, 2008)."]
Now, we looked at several things in regards to professional status because of what the literature says. So, for example, there's a lot of studies that are looking at what we call star scientists, okay? So, you think about it like, in Canada, these are the Canada Research Chairs or people that belong to the Royal Society of Canada. And so, in general, they attract disproportionately the largest pool of collaborators. It makes sense. They're famous. They will get people contacting them. They don't just have to contact other people when they want to collaborate. We also find that assistant professors collaborate more than others within their own universities rather than outside, of course, and full professors can do more international collaborations. And last but not least, of course, full professors tend to collaborate more with industry. Now, just to bring you into my little academic world, is to say that when you think about academics, assistant professors, majority of them don't have tenure, right? So, they tend, not just in collaborations, also in the kind of research they do, they tend to do sure things they want to do, they need to publish in order to get tenured, and hence their collaboration is with people and topics that they tend to do. So, we see this in different kinds of fields in regards to assistant professors.
[00:12:30 A slide is shown with the text:
"PI Characteristics – Clinical/Industry Experience:
Industry experience may negatively affect academic publication output (Lin & Bozeman, 2006).
Can also foster interdisciplinary collaborations (Van Rijnsoever & Hessels, 2011).
Bhullar et. al. (2017) indicates that prior collaborative experience with industry can generate a positive cycle: it equips academics with a better understanding of industrial challenges, facilitates access to new research opportunities and funding, and strengthens long-term research networks."]
Now, in this particular area of research that we looked at, it's a regenerative medicine. So, you have a lot of work with clinicians and with industry. And so, we try to see what the literature says about that and we find that industry experience actually will have less, it will have a negative effect on publications. And if you can think about it, in industry, it's more important to patent. We tend to publish before. So, a lot of times, people are much more careful in their publications before they publish. However, being a clinician or having an industry experience also fosters interdisciplinary collaboration because they have these unique outside-of-academia connections. And of course, we know that prior experience with industry can generate a positive cycle, right? So, you worked with industry, you keep working with industry. So, just as an example, if you think about Professor Drucker who did the initial patent that led to Ozempic, the initial relationship, it wasn't big income generating for U of T. However, they kept this relationship and came back to him once they developed, realized the importance of the drug in the diet area, and came back to the university to continue these relationships and help with the research, right? So, these are really, really important. So, what are we looking at specifically?
[00:14:03 A slide is shown with the text:
"Canada First Research Excellence Fund:
The Canada First Research Excellence Fund (CFREF) aims to boost the strengths of Canadian post-secondary institutions so that they can achieve global success in research areas that create long-term social and economic advantages for Canada.
CFREF invests approximately $200 million per year through a highly competitive peer review process to support the selected Canadian post-secondary institutions in turning their key strengths into world-leading capabilities.
Competitions – 2015, 2016, 2022."]
So, I'm going to tell you about the Canada First Research Excellence Fund, in short, CFREF. CFREF is a creation between the tri agencies, so this is SSHRC, NSERC, and CIHR. They don't do a lot of collaborations together. So, this is a very unique experience here where they created this fund and they invest about $200 million a year to create global success in the research area, and the first competition happened in 2015. That's the competition in which U of T, and I will tell you about the organization, received their grant. They repeated the competition in 2016 and then again in 2022.
[00:14:52 A slide is shown with the text:
"Research setting – Medicine by Design:
Established in 2015 with a $114M CFREF investment
Awarded >170 projects ~$75 million in funding
Central hub for innovative opportunities in regenerative medicine in Toronto Academic Health Sciences Network (TAHSN)
"Mission: To harness Toronto's extraordinary expertise across convergent disciplines and institutions to invent, develop, scale, and implement living therapies that will transform healthcare for Canada and the world."]
So, the organization that received funds from the first round of CFREF was Medicine by Design. It was established in 2015 using $114 million from CFREF. Just to tell you, in the last round, the highest award that CFREF gave was $200 million. So, we have expanded that over the years. Now, importantly, CFREF doesn't tell the winners how to use the funds. So, this is another thing I'm trying to look at in a different project but it's interesting on its own. So, CFREF, in a way, is giving the organizations somewhat of a clear hand to deal with research. Medicine by Design chose to invest the majority of the funds in research. So, here, you see $75 million was invested specifically in research projects. But in addition to that, they also funded 14 faculty appointments at the university. So, most of their funding did not go to the organization itself but actually to fund research. Medicine by Design sits on this collaboration between the universities in Toronto and the hospitals. It's called the Toronto Academic Health Science Network, TAHSN, and the mission of TAHSN and mission of Medicine by Design is to harness Toronto's extraordinary expertise across converging disciplines to develop and implement living therapies that will transform health care for Canada and the world. So, they were very engaged with their attempt to not just fund research but also see the commercialization outcomes from this research.
[00:16:35 A slide is shown with the text:
"Methodology:
Analyze activity of 168 MbD PIs funded 2015-2022 – N=1,344 person-years, Follow outcomes through 2023.
Analysis Part 1: Descriptive analysis – Descriptive analysis of researchers and social network trends.
Analysis Part 2: Econometric analysis of collaboration networks – Models for count data and unobserved heterogeneity.
Data on academic publications from 2005: Web of Science Core Collection
Additional sources for data on PI rank, affiliation, Tri-Council research funding, patents: Observatoire des Sciences et des Technologies (OST), MbD, web searches, professional social media"]
So, what did we do in this project? During the life of Medicine by Design, they funded 168 researchers. So, we looked at those researchers from 2015 to 2022. We looked at all their publications. At one point, we also looked at their grant applications and patents. We first do a general social network analysis. So, you'll get to see those really cool networks. And then, we also took the data from the social network analysis and compute it in an econometric analysis to tell us a little bit about the different variables on how much of it is the individual characteristics versus the geographical and the organizational ones. The data comes from publication data, so it's Web of Science. We look specifically at publications and new publications. So, we're looking at research, at Medicine by Design, as a treatment in time, if you'd like. So, what we see, we look at what happened in 2015, so before they started getting funding, and then in 2022, after years of funding. We also have a lag because it takes time for publications to come out. We also differentiate between regenerative medicine publications and just general publications. And as I said, we look at the geography, we look at the organization, and then we look at these different personal characteristics.
[00:18:11 A slide is shown with the title "Variables" and text that reads:
"Dependent: Cumulative annual count of ties for each MbD PI measured four ways:
- a) Publications/New Publications
- b) RM pubs/New RM Publications
- c) Geographic homophily
- d) Organizational foci
Independent :
- a) Female
- b) CRC Holder
- c) RSC Member
- d) Clinical Affiliation
- e) Industry experience
- f) Experience outside UofT
- g) Rank: Associate/Full"
"Control: MbD funding amount, Tri-council funding amount, Lagged values of dependent variables, Year fixed effects"]
RSC is the Royal Society of Canada and CRC is the Canada Research Chair. So, we're looking at all these gender, star scientist, and clinical industry affiliations, and of course, their rank as assistant, associate, full professors. So, now, we're going to go into the descriptive results.
[00:18:43 A slide is shown with the text: "Figure 1. Between MbD Researcher Collaboration Networks (2015 to 2022)". A graph of the PI collaboration network in 2015 is shown next to a graph of the PI collaboration network in 2022 by gender.]
Here, you see the first map that shows you the differences in the network between females and males. The females are the orange, I hope that's the same colour that you see, and the bluish one is the males. And you can see that in 2015, there were a few females that were part of this network but they were more in the periphery of the network. The size of the node that you see shows you how many collaborations they have. So, you can see that also they grew over time. The network itself is much more intense. Just to be clear here, we're just looking at the 168, because once we add all of their collaborations, you can't see collaborators, you can't see anything. So, this just shows you the specific PIs. And you can see that in 2022,women are taking central positions in this network and strong positions. So, they have intense and repeated collaborations with their collaborators.
[00:19:44 A slide is shown with the text: "Figure 1. Between MbD Researcher Collaboration Networks (2015 to 2022)". A graph of the PI collaboration network in 2015 is shown next to a graph of the PI collaboration network in 2022 by rank.]
Here, we're looking at the rank of professors. So, assistant professors are the orange ones, which don't exist even on the first map in 2015. But then, if you look at the map in 2022, they are part of the network, some of them have very central roles and large roles in the network, and you can see also the associates that were somewhat peripheral are taking also a central role in this network. So, if you want to think about it in the logic of research funding, the research funding allowed both assistant and associate professors to take a central role. They applied for grants, they received grants, and then they ran their projects and published.
[00:20:34 A slide is shown with a map that has green and red nodes on different locations below the title: "Does the Geography of MbD Researcher's Co-author Networks Change Over Time?" and next to text that reads:
"Legend:
Node size: Count of total ties, 2015-22
Green: more connections in 2022 than 2015
Red: fewer connections in 2022 than 2015
*Ontario excluded".]
Now, the geography is the part, because I'm a geographer, the geographies did not disappoint this time. So, I'm going to take a little bit of time to explain this. So, this is the map that shows the change over time, okay? So, this is a 2015 to 2022 map. The green nodes shows you a positive, so increase in collaborations, and the red is a decrease, okay? So, interestingly, the two big centres for regenerative medicine in the world are Boston and San Francisco which are still big on this map but they are negative. So, we have more new relationships in Canada over the U.S. And actually, we have those also internationally, okay?
[00:21:22 A slide is shown two additional international maps with various green and red nodes above a chart displaying a two-sample test of proportions: Canada-Canada, Canada-International, and Canada-USA.
"Two-sample test of proportions
2015 2022
Connection Category Freq. Prop. Freq. Prop. p-value
Canada-Canada 4,547 59.1% 8,324 65.9% 0.00
Canada-International 1,259 15.5% 2,530 20.0% 0.00
Canada-USA 1,882 24.5% 1,776 14.1% 0.00"
"Legend:
Node size: Count of total ties, 2015-22
Green: more connections in 2022 than 2015
Red: fewer connections in 2022 than 2015
*Ontario excluded".]
So, with these funds, which is somewhat what CFREF wanted to do and they are very happy with this result, I can tell you, we were able to increase our collaborations in Canada and internationally over collaboration in the U.S. Now, you're going to ask me this but I'm going to ask myself this, why? And I can tell you, the paper is in process. So, after a first round of reviewers, they wanted to know more information about this. So, we've done a few follow-up interviews, and part of the reason is we didn't have funding, as much funding, especially large grants, for regenerative medicine in the past as we had thanks to this grant. And so, most of these people were collaborators on U.S. grants and hence published. Some of them also graduated from the U.S., but that's what we think is the main reason for that. Okay, now, we computed everything and we put it into a regression model. I am not sharing the model with you but I will share the results.
[00:22:23 A slide is shown with the text: "Econometric results"]
[00:22:33 A slide is shown with the text:
"The Network:
Over the seven years the MbD program provided funding, its PIs strengthened collaboration within academia and between academics and clinicians in affiliated hospitals.
Clinical affiliation, star scientists status, full professorship, and prior experience are all significant predictors of expanded co-authorship networks, reinforcing the importance of diverse backgrounds in fostering collaboration.
Within regenerative medicine specifically, increased collaboration was most evident among female PIs, star scientists, and full professors."
Connection Category 2015 2022
Freq. Prop. Freq. Prop. p-value
University-Hospital 303 17.8% 2,108 22.0% 0.00
University - Industry 105 6.2% 149 1.6% 1.00
University-University 1,125 66.1% 6,426 67.1% 0.20" ]
I think one of the interesting results that we have is that over seven years of this program, we can see more collaborations between university and hospitals instead of just university with university, okay? Sorry, instead of university-industry relationships, but we increased these relationships with hospital, with clinicians, okay? Well, we also find if you look at the specific characteristics of the PIs, is that having a clinical affiliation, so I don't know if how many of you know but most of the clinicians at the hospitals have a U of T affiliation, okay? That doesn't mean they are at the university. So, if we're talking about the organization that we work with, this collaboration actually increased the relationship with clinicians, okay? So, having a clinical affiliation, having a star scientist affiliation, so Royal Society or Canada Research Chairs and full professors with prior experience outside academia, are all significant predictors to expanding the networks, okay? If you look specifically at papers that were published in regenerative medicine, because you have to remember some of these people continue to publish in other fields, not just in regenerative medicine, but specifically for regenerative medicine, the increased collaboration was most evident among female PIs, star scientists, and full professors. Now, this is a great way to show, we always tend to say women don't collaborate enough, well, here, in this particular field, with the support of the policy, they were able to increase their collaborations.
[00:24:31 A slide is shown with the text:
"Our results underscore the role of both homophily and organization foci:
In terms of geography, we find that MbD funding expanded collaborations within Canada and internationally, while reducing reliance on pre-existing U.S. networks.
Except for clinicians and Canada Research Chair (CHC) holders, most groups expanded collaborations domestically and abroad.
Organizational foci reveal that while clinicians and full professors engaged across hospitals, universities, and industry, only PIs with prior industry experience or RSC status collaborated beyond academia."]
Now, what does it mean, there are questions about geography and organizational foci? So, in terms of geography, as I showed you from the map, MBD's funding expanded collaboration within Canada but also internationally while it reduced pre-existing U.S. networks. And except for clinicians and Canada Research Chairs, most groups expanded their collaboration both domestically and abroad. But specifically for clinicians and full professors, they engaged across hospitals, universities, and industries, okay? And of course, the PIs that have both industry experience or Royal Society of Canada, which means that they are highly known internationally, collaborated beyond academia.
[00:25:37 A slide is shown with the text:
"Conclusion:
MbD regenerative medicine PIs not only deepened co-authorship ties among themselves but also reshaped the structure and reach of their networks across demographic, organizational, and geographic dimensions.
Highlight the critical role of strategic funding in fostering interdisciplinary and cross-institutional collaborations in emerging fields.
Collaboration remains bounded by organizational foci and national research ecosystems, suggesting that further interventions may be required to expand the networks."]
What does that mean for us when we think about policy and funding research, okay? So, the regenerative medicine PIs not just deepen their ties between themselves in that small networks, but actually, if you'd like, they reshape their network. And I think importantly for us and importantly for U.S. policymakers thinking about how policy impacts for the benefit of Canada, you can see how this kind of funding helped increase the collaborations in Canada but it means also that we created specialization in regenerative medicine all across Canada. The research also highlights the critical role of strategic funding, right? So, especially when you think about cases of interdisciplinarity and cross-institutional collaborators, I can tell you as somebody who works on multidisciplinary research, it's very difficult to get funded in Canada. We tend to fund political science, geography, management. We don't do these cross-discipline funding. So, when there is funding like that, you can see the importance of it because of these relationships that it creates and it helps us create new knowledge in the field that was just started. And of course, somewhat, we have this organizational foci, right? The institutions, the clinicians, and the hospitals, but it means that if we want to expand collaborations, if we want to make sure that new knowledge comes in, these kinds of tools are really important when we think about creating new fields and multidisciplinary collaborations.
[00:27:37 A slide is shown with the text:
"Munk School of Global Affairs & Public Policy Innovation Policy Lab – University of Toronto"
"Thank you!"
"Shiri.Breznitz@utoronto.ca"
"We acknowledge financial support for this work from Medicine by Design CFREF. We are grateful to the research assistance of Mauricio Sepulveda, John Wang, Que Kanh Luong, Hannah Wu, Jenna Muschong, and Catherine Yeh. We extend thanks as well to Michael Sefton and Allison Brown for feedback on earlier versions of this work."]
And that's it, thank you.
Cheryl Engler Wadasinghe: Thank you, Shiri, for your presentation. We're ready now to move on to our question and answer session. So, just before we begin, I'd just like to remind everyone that you can submit your questions by clicking on the little chat bubble icon that's located at the top right hand corner of your screen. And Shiri, if it's okay with you, maybe I'll start with a question from me (laughs). Your study results, it's a fascinating area of research to look at the cross-sectoral, cross-interdisciplinary area and especially in areas of research that are new and emerging. Before I came to the School, I was working in forensic analysis to advance criminal investigations, and I got in that work to see first-hand how even inter-laboratory collaboration can assist in advancing implementation of new technology, which generated for us more probative results and that had benefits for the federal courts. It's delicate and time-consuming though to do that work, and I was just interested to ask you what sparked your interest in studying how researchers and educational institutions and hospitals, clinics, and industry collaborate and build these networks. What was the seed that grew for you from that?
Shiri Breznitz: Yeah, so I've worked on the role of universities in economic development for a long time and have done work on accelerators, incubators. But coming here, especially when you realize… physically when you see the hospitals that are one next to the other but also close to the university. So, I had a chance to be involved with Medicine by Design and learn about their work. And so, I was really curious to try and understand how these collaborations work. We know that collaborations, strong collaborations, have a positive impact on innovation, right? So, trying to understand how the hospitals and universities collaborate. And by the way, if you hear people talk about, we did some focus groups, and people think, it's natural, they're also academics, the clinicians also do research, but it was very difficult for them and it took a while to do this and collaborate to the point that now the tech transfer offices collaborate together, they submit grants together, but it wasn't so natural, even though they supposedly had that umbrella, TAHSN, existing in the past. So, for me, this was like a signal. I wanted to try and understand what happened there.
Cheryl Engler Wadasinghe: Right, right, yeah, how it kind of grew and came together. Thank you. I do have another question for you. How do collaboration networks compare to more traditional innovation measures? So, you did speak about that a little bit at the very beginning, but could you elaborate a little bit more on that?
Shiri Breznitz: Yeah. So, with collaboration networks, we can see all of these. So, there are different ways to show the collaborations themselves, right? So, you see the network itself and you see the output, if you'd like. So, we are using the network to do an analysis on the publications, on the patents, and on the grant applications, all of them together.
Cheryl Engler Wadasinghe: And yeah, that's the additional, I see, yeah. I'm curious as well to ask about the funding model. You had mentioned that one characteristic of the CFREF is that it isn't prescribed as much as other models might be. Could you comment a little bit on the nature of that model, if there are other characteristics about it that you felt made a difference in terms of how this industry or this sector of research has been supported?
Shiri Breznitz: Yeah. So, as I said, we've done several focus groups. CFREF was included but also other CFREFs in Canada were included, and it's very interesting because each of them chose different ways. Some of them just created an institute and encouraged collaborations, they did events and workshops and training, where others funded research. So, I think their point was because some of these were more traditional areas, some of them were new, they didn't want to prescribe that way. They wanted their researchers and their organizations to find the best way to promote research. So, their point of view is, we are trying to promote research but we are sitting in policy, not necessarily knowing the specific research. For example, one of the areas is called oceans. And so, you have researchers from so many different disciplines that would not have any other way met unless you had this opportunity, right? So, if you look at the studies on research collaboration, some of them will say, okay, this doesn't really give us a prescribed way. On the other hand, when you talk to these organizations, some of them tell you that because it was so open, they could find the best way, to their specific discipline, their specific organization, how to collaborate and encourage collaborations.
Cheryl Engler Wadasinghe: Right, and then have that discretion, I guess, to set up that structure more themselves in terms of their priorities. That's very interesting. Another question has come up from participants. "Aside from the gender lens, was there consideration of other equity characteristics in the analysis in the PI or networks?"
Shiri Breznitz: Yeah. So, we looked at their gender. We looked at their rank, so assistant, associate, full. We looked at whether they are star scientists, so the Royal Society of Canada and the Canada Research Chair. We looked at whether they had experience outside academia. We also looked at, there was one more thing, clinical, so whether they have clinical experience or not clinical experience, yeah.
Cheryl Engler Wadasinghe: And I think the question, if I understand it correctly, is also asking about other types of equity characteristics, and I guess the typical ones would be age or ethnic background or race and that kind of characteristic.
Shiri Breznitz: Yeah, no, no, we don't have that information. So, we were looking very specifically, assuming all of them are in Canada, in Toronto, in a way, you're controlling for some of it. Of course, we know we are probably one of the most diverse places in the world. But still, you assume that at least academically, they have the same culture. So, that's part of it. And then, of course, you have the organizational foci and the geographical distance.
Cheryl Engler Wadasinghe: Right, and I guess it could be, with the information, a future area that could be included.
Shiri Breznitz: Yeah, and I would say age, in some way, you can also deduct from their rank, right? To a degree.
Cheryl Engler Wadasinghe: That's true to a degree, yeah, yeah. All right, our next question, "Your presentation brings to mind the bibliometric research done as well as the Web of Science at the University of Ottawa, which is wonderful to see. Do we think that we would see similar results across other multidisciplinary fields of study?" What are your thoughts on that?
Shiri Breznitz: That's an excellent question. Yeah, it's an excellent question. So, I was trying to continue this work and we still might, I don't know yet, but part of the idea is to look at these other CFREFs. Some of them are more multidisciplinary than others. So, it's a great case to look and study but I would expect to see differences. Here, you have a lot of clinicians involved. And so, I would think that we will see differences, especially since the result shows that being a clinician is important and the relationship with the hospital is important. So, I would assume that we will find differences across different fields.
Cheryl Engler Wadasinghe: Right. All right, our next question is, "Is there such a thing as too much collaboration? Can collaboration ever slow down innovation?
Shiri Breznitz: Definitely. So, if you look at studies on social networks, what happens in many places, you get what we call a network locked in, where we are so much into collaborating with the same people all the time, or when you think about it, it doesn't have to be five people but it can be a regional thing. So, for example, Cambridge, U.K. had a lot of these issues where it was very hard for new people to penetrate. Even though they were there sometimes physically, they weren't able to penetrate the network. And so, you create a situation where, as I explained, if you want new knowledge and you want to develop new areas, you need the knowledge to come from the outside. So, when you have a network locked in, you don't see that. And so, this is something that has been studied in research, not by me but there are articles. And if people want to write, I'm happy to share those with you.
Cheryl Engler Wadasinghe: That's very interesting. So, one of the things I'm taking away then is it's not collaboration. We can't look at collaboration as just one thing. You need to look at what's going on in the collaboration, so whether it's well-established and kind of has strict kind of governance around it or practices and how does that structure then assure itself that it's got some flexibility to be able to bring in new thinkers and different types of knowledge. Thank you. All right, I can move on to the next question. "Based on your work on social networks and innovation, what makes a collaboration network effective and what are the common barriers that limit its impact?"
Shiri Breznitz: Sorry, can you repeat the question?
Cheryl Engler Wadasinghe: Of course. "Based on your work on social networks and innovation, what makes a collaboration network effective and what are the common barriers that limit its impact?" So, a little bit similar to what we were discussing in the context of structure, but are there other factors?
Shiri Breznitz: Yeah. So, for example, when we looked at startups, remember with my example from Georgia Tech with the accelerators working together, so when we talk about culture, sometimes it's language, direct language but also the academic language or the professional language. A lot of times, especially when you think about academics, people talk about the issues of academia and industry collaboration is this shared language. It's because we tend to think we have so much time to do our work, and the industry wants it yesterday, right? So, this is kind of a cultural thing. So, for social network success, so beyond the things that I talked over there, I provided over there, yes, age is definitely an issue but part of the age is also being established in a field or in the network, if you'd like, there is culture, both, as I said, academic but also ethnic culture. So, I was part of a research group in MIT a long time ago where we had a meeting of Americans and Norwegians. And before the meeting started, the director said, okay, I want to remind people that we have Norwegians and Americans, and Norwegians, I please ask you to ask and please participate, and the Americans, please remember that the Norwegians tend not to, so leave them space to participate, right? And sometimes, these small things are really important, yeah.
Cheryl Engler Wadasinghe: Right, right, to keep in mind those, yeah. That's very interesting.
Shiri Breznitz: Differences, yeah.
Cheryl Engler Wadasinghe: Differences, yeah. And in a way then, a collaboration that works or what exists and what you want to grow, they'll have their own kind of personalities or characteristics based on who's participating and who you, I guess, want to draw into the conversation.
Shiri Breznitz: So, it's the people themselves. It's the environment. So, it's the location, location definitely matters. But also, it's the general culture and whether you have mechanisms. In this case, we had funding mechanisms that helped promote the network, but you can think about events. So, it's a part that I couldn't present to you yet but we are doing an analysis on the events that Medicine by Design created to see if people who participate in the event have more collaborations with people that they didn't collaborate before, that maybe they met at the events. So, all of these, it goes the person, if you'd like, the location, the environment that they're in, and then the culture.
Cheryl Engler Wadasinghe: Right, right. Thank you. Onto our next question, was visibility or social media dissemination frequency by researchers considered in your study and in your… sorry, I have to start this again. "Was social media dissemination frequency by researchers considered in your study and how this influenced collaboration opportunities or productivity?
Shiri Breznitz: No, we didn't look at social media at all. This is really based on academic publications, patents, and grants.
Cheryl Engler Wadasinghe: Right, okay. If the public service wanted to apply your findings to its own institutional reality, what's one practical step that you would suggest that we take to better support or incentivize collaboration across departments? So, that's within the federal government.
Shiri Breznitz: So, we can't control a lot of the personal characteristics, right? But if we want to promote junior versus senior collaborations and across disciplines, I know it's different because in Canada, people tend to move, so you have people who worked in different areas, but maybe create some kind of champions of people who did work in different areas and then they understand the culture of that office or department or understand even the basic knowledge of the kind of work that you do in order to create these collaborations. There's some that probably collaborate on a regular basis, right? And others less, so trying to find those, if you'd like, holes and see where you can connect. A lot of times in academia, we try to put champions for these connections.
Cheryl Engler Wadasinghe: Right.
Shiri Breznitz: And when you think about entrepreneurships, a lot of times we have entrepreneurs in residence, they make that connection with industry because they can speak both languages, if you'd like, so something similar. And of course, when you have cross-disciplinary or cross-department events, I think something like the Canada School of Public Service is a great opportunity, right? When you go to these meetings, you meet people from different departments that you probably would not meet otherwise, right? And you learn about the work and the kind of things they do and how they do work in a different way, which is really important when you think about new ways and methods, learning about a different way of analysis that a different department is using. You might decide to adopt it, you might decide to learn from it and adopt certain things, but if you don't have those connections, you will never be aware of these relationships and knowledge.
Cheryl Engler Wadasinghe: Right, right, right. Well, something I really am taking away from a comment that you made earlier was the idea that if you look at the existing collaborative networks that we have, there are kind of portfolios of organizations or federal departments and agencies that are very much in the habit of working together but I don't know how much time we allocate or how much time, a lot of people say time we have, to go to consider, well, what are other agencies and departments doing in this field? Even though, strictly speaking, it's tangentially related to where we're going. It's definitely something to keep in mind in terms of, especially as you mentioned, these emerging issues.
Shiri Breznitz: This is why creating these kinds of organizations, so the Canada School of Public Service is a good example, but in the same way, Medicine by Design is, without this organization, you don't have those at all because you come, you do your work, you have some things to accomplish, you don't automatically and you don't always know also what people are working on. But if you can have a body that creates these collaborations or these opportunities at least to collaborate or to learn more, then we already have done half the work to bring people together.
Cheryl Engler Wadasinghe: Right, right. And in terms of after the funding was confirmed, either in 2015 or subsequent years, what kind of structure was put in place to help bring ideas together and make decisions about the allocation of the funding? I know that's more a question of how the model worked once the funding was approved and less about the networks, but do you have confirmation of that?
Shiri Breznitz: That's actually in the paper, yes. So, they actually had a very strict review process. So, they have a board but they also have a scientific board. So, I believe they had four calls for funding over the years. For each one, the applications went to international reviewers. And then, there was a committee that reviewed those and made a recommendation to the board to decide who to fund. So, it was very academically done, using the, I believe, CIHR ranking rubric.
Cheryl Engler Wadasinghe: Okay, that's very interesting because in a model that's less structured, I guess, on the other side, on the recipient side, you'd have to put something in place to be able to have a transparent process.
Shiri Breznitz: Yeah, so they did it for themselves.
Cheryl Engler Wadasinghe: Yeah, exactly, a transparent process.
Shiri Breznitz: And actually, CFREF required them to report every year. They had to report where the funding went, who they funded. It's all very transparent on how they spent their money over the years. I think there was several reviews that asked them to make changes, and not just them, the other CFREF. So, CFREF the organization was highly involved within those projects, in the review process also, over the time of the grant.
Cheryl Engler Wadasinghe: Okay, okay, that's very interesting. I have another couple questions for you. I'm just keeping my eye on the time. Is there a rule specifically, like are any of the partners in this discussion, in this collaborative model, challenged more than others? Or for example, non-academic actors, is the way that they're contributing to this changing or do you see that changing as a result of this type of funding model?
Shiri Breznitz: I'm not sure I understand the question because everybody, in a way, is an academic. As I said, all of these clinicians are U of T. They have a U of T affiliation, so TAHSN. So, CFREF was very specific that the funding needs to go to this group. They actually found it very difficult. That's why they were happy to see that the collaboration went beyond, because they couldn't fund people outside this group. This is the requirement of CFREF. So, they're all part of this particular funding team.
Cheryl Engler Wadasinghe: Okay, interesting. I'm just trying to think of the right words for my next question. Based on your research and what we're hearing about the importance of being able to fund Canadian solutions where we might have gone outside the country previously, how applicable is this type of model to some of those sectors where we're trying to grow our innovation and develop Canadian solutions? And I guess part of it is challenging because some of the solutions we'll be looking to build already exist as opposed to in regenerative medicine, it was a field that was growing globally at the time, and so, if you will, the race was on for everyone, yeah.
Shiri Breznitz: Yeah, but we are quite late by the way. So, in the U.S, regenerative medicine was funded as early as 1992. So, there are several regenerative medicine institutions. We were quite late for it, and that's why I think we saw more relationships in the past in the U.S., because people were on grants with U.S. collaborators. So, especially in the sciences, I don't know all the funding mechanisms but we do have something called NFRF, so new frontier research. So, we do have these funding programs. NFRF, I believe, is every two years only but I am not super familiar with it. We don't have a lot of funding that's directed to cross-disciplinary collaborations to create something new. Most of the time when we think about funding, this is what it traditionally is. It's very much committees that are based on specific scholarly fields, so physics, math, traditional like that.
Cheryl Engler Wadasinghe: So, in a way then, being able to apply this or replicate it could be very interesting for breaking down some of those, I guess, silos, you could say, and creating more opportunities for interdisciplinary work.
Shiri Breznitz: Definitely, definitely. I think this is one of the things that the U.S. has that strength because they traditionally had these kinds of funding.
Cheryl Engler Wadasinghe: I see.
Shiri Breznitz: But it's not that we don't totally have those, they're just more limited than what I would like to see.
Cheryl Engler Wadasinghe: Right, what we could be doing, yeah, okay, okay. Well, thank you very much. I have one more question as we wrap up. What key lessons from your research should we be thinking about as public servants working in science and innovation and research funding? I mean, we touched a little bit on the idea of replicating the way that CFREF is set up but are there other things that you would suggest to us in terms of collaboration in emerging fields?
Shiri Breznitz: Yeah, I think, first of all, it's difficult. It's nice now to sit back and for me to review something that was done years ago, almost 11 years ago, and say this was amazing. It's very hard to foresee the future, right? And part of the reason why we keep funding basic research is because a lot of the most disruptive technologies in the world came from funding basic research. We didn't plan. Again, Ozempic is a great example. The first invention was research for diabetes. It wasn't for weight loss, right?
Cheryl Engler Wadasinghe: Yeah.
Shiri Breznitz: So, if we didn't go that route, if we were looking for a solution for weight loss, we probably wouldn't have gotten to this. So, there is a huge advantage of funding basic research and I'm a true believer in that, and I think there's room for collaborations and room to support this kind of work as part of the general funding for basic research, so making sure that all of these avenues are open because we never know where this new knowledge is going to come from.
Cheryl Engler Wadasinghe: Right, right, thank you. Thank you very much, Shiri, for the presentation. On behalf of the School, I would like to thank you very much for the presentation and for the conversation that we had. It was very enjoyable. I'd like to thank all of you in our audience from across the country for being a part of today's discussion, and I hope you found this conversation inspiring. I'd like to encourage our audience to visit the Canada School of Public Service website to keep up to date and register for future learning opportunities such as this one, and I'd just like to take this opportunity. Shiri, to thank you again. I hope everyone has a wonderful rest of their day.
Shiri Breznitz: Thank you.
[00:54:04 The CSPS logo appears on screen, then the URL canada.ca/school.]
[00:54:08 The Government of Canada logo appears on screen.]
