Field-Driven Research Synthesis: Understanding the Character of Innovative Interdisciplinary Exploratory Research

Exploration and Discovery Across New Frontiers

Engaging with partners to generate ideas across disciplines: Roxanne (AI), Sheryl (Accessibility and Learning), Marjorie (Simulations), and Lorna (Accessibility and Learning)

Engaging in exploratory research allowed PIs to convene partners and generate ideas across disciplines by a) fostering collaboration with partners from different fields, b) identifying knowledge gaps, and c) developing ambitious goals and innovative solutions. Below we share specific quotes to illustrate these activities.

The NSF EXPs specifically called for and supported interdisciplinary approaches to exploratory research. For example, Roxanne described how her AI project, Exploring AI-Enhanced Electronic Design Process Logs, brought technologists, practitioners, and subject matter experts together:

I thought it was a really special opportunity in terms of the collaborations, working across disciplines, and having a solicitation that’s specifically set up for multidisciplinary and interdisciplinary work.

For many of the PIs we interviewed, including Roxanne, the NSF funding afforded opportunities for interdisciplinary engagement. This collaborative engagement across disciplines bridged knowledge gaps, as Sheryl explained in her Accessibility and Learning project, Designing STEM Learning Environments for Individuals with Disabilities:

I think it can provide a foundation to go to the next step. And I also think it’s a type of research where we can get multiple stakeholders together. In my experience, you need to take the time to make them comfortable with each other. I consider myself to be kind of a social engineer. You start with, “Well, they don’t talk to each other, so we’re going to bring them together.”

Sheryl’s exploratory project convened partners from diverse disciplines with contrasting perspectives who do not normally engage each other: individuals with disabilities, cyberlearning technology and pedagogy researchers, computing faculty and doctoral students, cyberlearning instructors and designers, leaders of K-12 and postsecondary cyberlearning projects, and IT accessibility experts. This interdisciplinary approach to interdisciplinary, exploratory research helped identify needs and gaps in creating accessible technology in this community and set the stage for future research and development.

Interdisciplinary approaches in exploratory research allow project teams to both identify knowledge gaps that might otherwise have been missed and develop ambitious goals.

Discussing the Simulation project Exploring Social Learning in Collaborative Augmented Reality, Marjorie explains how interdisciplinary collaboration allowed her team to identify and focus on knowledge gaps within emerging technologies research:

That’s what’s so great about NSF work: We’re able to identify gaps. We don’t have to solve every research issue right at that moment. We’re able to identify gaps and then fill them in as we go forward.

Finally, collaborating across disciplines supported the development of ambitious goals and arrived at innovative solutions, as Lorna described in her Accessibility and Learning project, Implementing and Testing Signing Avatars & Immersive Learning:

Having the opportunity to try to chase those goals that we have, even though they’re high, and even though they require a lot of different areas of expertise that are not always seamlessly integrated, has allowed us to make a lot of progress with this project. For example, it is not extremely typical that a professor of educational neuroscience like me would have a postdoctoral researcher whose background is solely in computer science and deep learning. So, we are both taking a leap of faith that we can work together across disciplines to make something. And by reaching across disciplines, we are able to pull together different pieces into one cohesive project.

Taking a leap of faith and chasing ambitious goals in exploratory research, as Lorna shared, requires engaging interdisciplinary perspectives and expertise.

Exploring emerging technological possibilities: Rebecca and Colleen (Virtual and Augmented Reality), Nikolas (AI), Erin (Collaborative Learning), and Lorna (Accessibility and Learning)

Interdisciplinary, exploratory research provided avenues for PIs to explore emerging technological possibilities by a) funding passion projects in understudied and semi-structured spaces, b) affording freedom to explore possible and unimagined futures, and c) enabling research considered high risk or unorthodox.

Our interviews revealed strong beliefs that engaging in interdisciplinary, exploratory research creates a space for passion projects in underappreciated and understudied spaces, as Rebecca, co-PI with Colleen for the Augmented Reality project Combining Smartphone Light Detection and Ranging with Augmented Reality, explains:

This particular project is not through my current affiliation, nor was it through my previous one. This comes out of my teaching experience as a high school teacher, and it comes from my husband who is a software developer. This is very much a passion project. So, I see exploratory research as a really appropriate place for passion projects that can have a lot of impact in a research space that is underappreciated, which, you know, is high school physics teaching.

Engaging with data in semi-structured and unstructured problem spaces also gave PIs the flexibility to focus on unanswered, novel, and reframed research questions compared to more established lines of research. This allowed for speculation and for new ideas to develop, as Colleen commented:

I think the exploratory research allows you to know more than just a snapshot at the beginning and a snapshot at the end. How do things unfold? We’re doing something that’s never been done in a classroom. We don’t know! We have maybe a mental model of how we think things are going to play out, but we don’t really know. You need a relatively unstructured problem space in which to develop something new.

Given the understudied nature of interdisciplinary, exploratory research, it was important to the PIs that NSF funding opportunities not be tied to conventional or established research lineages. For example, during Nikolas’s AI project, Supporting Designers in Learning to Co-create with AI, co-creating interactions with AI were in their infancy, and the quality of AI-generated content was not as advanced as it would soon become:

The thing that we think was innovative and exploratory was that, at the time, there weren’t a lot of generative AI co-creative interactions. There was some work in the space, but the level of excitement and the functioning of the AI, the actual quality of what was generated, wasn’t as good…in many ways it was exploratory to figure out that “if this is coming, we think it’s coming soon.”

Rather than “build better AI, make it work better.” That’s what AI companies are doing. They’re great at that. That is not what our group focuses on. We focus on the human interaction side, and we focus in many ways on making the human better…For example, these agents that help you think through things, these meta-cognitive agents, work. We will make the person better. And every time the AI gets better, hopefully the whole interaction gets better.

Anticipating the imminent advancements in AI, along with the focus on enhancing human interactions with AI, made Nikolas’s work pioneering. As a result, many PIs, including Nikolas, felt that engaging in interdisciplinary, exploratory research provided freedom to envision potential futures beyond the limitations of current technologies.

Erin expressed this sense of expanded imagination and opportunity while discussing the Collaborative Learning project Improving Student Help-Giving with Ubiquitous Collaboration Support Technology:

For all of my career up until now, my research has been entirely exploratory, and I see that as the scientific role I am filling. What I am interested in is presenting ideas to people of possible futures… With the hope of then encouraging people to take [ideas] up in ways that produce more effective learning technologies, but also aspirational technologies. Technologies that are more agentic in the kinds of agency they enable in learners. Or more empowering, more context sensitive, more responsive. So presenting a future of what technology could be in contrast to what it is right now… To re-imagine better futures in a way that I don’t think you could if you were just doing safer, more incremental, research.

Erin, like many of our interviewees, was passionate about exploring possible and re-imagined futures—particularly for aspirational technologies in under-researched areas. However, this freedom also presented tensions between embracing the open-ended nature of interdisciplinary, exploratory research and managing the high stakes of developing emerging technologies for learners.

Erin explained:

In this space, you have something that is very difficult, potentially impossible, to build. And that’s a technical challenge. And then you have this need to deeply understand the educational context, the individual learners, the factors of interest before finalizing a technology design… There’s all these challenges around negotiating between the practical needs of whatever context you’re in and the scientific needs of the project that’s maybe reaching 5 or 10 years into the future.

For some PIs, including Erin, engaging in exploratory approaches presented challenges between balancing freedom and high stakes in ways that were intimately tied to the needs and vulnerabilities of the learner populations that the PIs were trying to serve:

So I feel like there’s all these tensions that happen that require negotiation as you go from, “I need to deeply understand this” versus “I need to produce something,” to “this really needs to work very well in this particular context”—which is a very serious concern especially if you’re dealing with members of minoritized groups. A lot of my work has been very context embedded, which makes the stakes very high for something that’s also risky and that has just been this constant challenge.

Finally, the ability to explore new ideas and possibilities enabled PIs to focus on next-level research of emerging technologies considered to be high risk or unorthodox, as Lorna shared with us:

The NSF took a chance on us. I don’t just say that because I’m reading between the lines, but literally. In 2018, we submitted the first version of this grant proposal, and we received reviews that included comments like “overly ambitious.” Thankfully we had a very supportive program officer who saw the potential in this work and recommended that we be funded for an EAGER award. It’s a special mechanism at the NSF that allows program officers to provide partial funding for particularly high-risk projects… There was a big question of “Can it work? Can we build a high-quality Signing Avatar, put her in a virtual environment, and use her to teach ASL?”… We were able to show that the idea can take off and can work on a proof-of-concept level, and then we were able to submit a full proposal in 2020 and receive funding.

Notably, the program officer recognized the ambitious potential of Lorna’s project. At least several lines of underfunded research might not have come to light as NSF EXPs were it not for alternative channels such as EAGER—a funding mechanism that supports high-risk/high-reward interdisciplinary, exploratory research involving radically different approaches and novel interdisciplinary perspectives.

For other PIs, eschewing more established approaches to research—such as large-scale randomized studies using statistical analysis—was critical to the deep exploration of emerging technologies for learning, as Colleen explains:

Schools are unfortunately driven by statistics. And the mean does not tell you the whole story. There’s a whole lot of data that does not sit at the mean. There are a whole lot of students, different kinds of students, who learn in different ways. And we get to see that, and we get to play with that because we have the flexibility to say, “Okay, well, we thought we were going to use it in this setting, but let’s try it in this other setting.” And we’ve been able to do more than we originally planned.

Rebecca and Colleen’s use of exploratory approaches to analyze data that “does not sit at the mean” was critical to better understanding the impact of student variability and context in their research. Moreover, PIs’ focus on developing emerging technologies to support “different kinds of students, who learn in different ways” required engaging in equitable co-design practices, which we discuss in the next section.

Back to topNext section