Beyond Traditional Boundaries: How Big Team Science Can Dramatically Improve SIL Research

By Drs Hans Rocha IJzerman , Elizabeth Ghogomu , Samia Akhter-Khan , and Vivian Welch (she/her)

Social isolation and loneliness (SIL) research is facing a credibility crisis, largely driven by low-quality studies, insufficient sample sizes, and an overreliance on small-scale interventions that lack replicability. As highlighted in recent evidence gap maps (Welch et al., 2023, 2024), over 70% of systematic reviews are of critically low quality, creating an urgent need for transformation in how we conduct this research.

One solution lies in transforming how we approach SIL research by embracing Big Team Science (BTS) and developing a collaborative research network that rapidly tests, refines, and scales interventions through large, coordinated teams. These frameworks can not only raise the quality of research but also make interventions more effective, globally representative, and scalable. However, existing efforts in large-scale research collaboration, like BTS, often fall short by focusing on conservative ideas and underrepresenting researchers and populations from low- and middle-income countries (LMICs). This post will discuss how we can adapt Big Team Science models to overcome these challenges and propose a collaborative framework to fast-track meaningful change in SIL interventions.

The Need for Big Team Science in SIL Research

Big Team Science (BTS) involves coordinating multiple labs and research teams across different countries and institutions to conduct large-scale studies. BTS has gained prominence in psychology and other fields as a way to address issues of replicability, generalizability, and study power. With SIL research facing similar challenges, BTS can provide solutions by pooling resources, combining diverse data sources, and integrating expertise across geographies. The Psychological Science Accelerator (PSA), a network of over 700 researchers, has shown how this model can be scaled for collaborative science.

However, the problem with current BTS efforts, including the PSA, is that they often lean towards conservative research questions and focus primarily on high-income country (HIC) populations (Forscher et al., 2021). This can limit innovation and prevent research from addressing the unique needs of underrepresented populations. In SIL research, which affects people across all age groups but particularly impacts older adults and marginalized groups, it is essential that research collaborations reflect the diversity of those affected.

Building a Collaborative SIL Research Network: How It Works

A coordinated research network would provide a focused approach to addressing the quality and scalability issues identified in the evidence gap maps. By coordinating the efforts of multiple research teams, policymakers, practitioners, and funders, this network could systematically address the methodological and geographic gaps in current SIL research.

The Untapped Potential for Large-Scale Collaboration

A recent survey one of us conducted for the Global Initiative on Loneliness and Connection reveals substantial untapped capacity for this kind of collaborative work. With over 1,300 volunteer hours available monthly from researchers across career stages and geographic regions, there is significant potential to implement large-scale studies. This volunteer base spans multiple continents and includes researchers from both high-income countries and LMICs, creating opportunities for truly global research that represents different cultural contexts and perspectives on social connection. The interdisciplinary mix of volunteers - from psychology and neuroscience to public health and social sciences - provides the diverse expertise needed to tackle complex social connection challenges through Big Team Science approaches.

The network would work across three pillars:

1. Large-Scale Testing and Innovation: By coordinating the efforts of multiple research teams globally, the network would enable rapid testing of interventions across different contexts and populations. SIL interventions that have shown promise in smaller studies could be replicated and adapted to diverse cultural and social settings, particularly in LMICs. This collaborative effort would increase the statistical power of studies and allow for meaningful subgroup analyses, addressing one of the main shortcomings identified in current research.

Example: The UK Biobank, a biomedical accelerator, has enabled large-scale research by pooling data from half a million participants. A similar model could be applied to SIL research, allowing for data collection on interventions across age groups, health conditions, and socioeconomic statuses (Sudlow et al., 2015).

1. Focus on High-Risk, Innovative Research: Existing BTS efforts often default to low-risk, conservative studies to ensure broad participation and minimize conflicts. The network would need to change this dynamic by actively encouraging high-risk, high-reward research—interventions that may be innovative but come with uncertainties. This could include studying societal-level interventions that address structural factors like housing, transportation, or community design, which are currently understudied according to the evidence gap maps.

Example: The Open Innovation Framework, used in the pharmaceutical industry, encourages high-risk innovation by offering tiered funding for different levels of research risk. Such a framework could be applied to SIL research, with specific tiers dedicated to interventions that target marginalized or understudied populations (Allarakhia, 2018).

1. Commitment to Data Sharing and Registered Reports: For transparency and reproducibility, the network would adopt a policy where all studies are required to follow registered report guidelines and commit to data sharing. This addresses the critical need for higher quality research identified in the evidence gap maps. Registered reports ensure that study designs are peer-reviewed before data collection, minimizing the risks of p-hacking and selective reporting. Meanwhile, open data repositories would allow other researchers to validate findings and perform secondary analyses

Example: The Open Science Framework (OSF) offers templates to assist researchers in managing projects transparently. Through similar initiatives, the network could create shared platforms for SIL researchers to collaborate, share data, and enhance reproducibility (Nosek et al., 2018).

Addressing LMIC Representation and Global Generalizability

Despite the benefits of BTS, one of the most pressing issues is the underrepresentation of LMIC researchers and samples in large-scale studies (Ghai, Forscher & Chuan-Peng, 2023). As the evidence gap maps show, the majority of current research comes from high-income countries, creating significant knowledge gaps about intervention effectiveness in different cultural and socioeconomic contexts. The network would actively counteract these barriers by ensuring that LMIC researchers are not just participants but leaders in research efforts.

Key strategies to include LMICs:

• Dedicated Funding Streams: Establishing dedicated funding streams for LMIC-led projects within the network can incentivize the inclusion of diverse perspectives and ensure that interventions are adaptable to countries outside the Global North by local researchers.
• Training and Capacity Building: Implement training programs that equip LMIC researchers with the tools needed to lead and contribute to large-scale interventions. Capacity-building initiatives can focus on research design, data management, and open science practices.
• Collaborative Leadership Structures: Promote a collaborative leadership structure within the network, where LMIC researchers co-lead projects alongside HIC partners with fair authorship representation (Abimbola, 2024). This ensures that LMICs have a stake in shaping the research agenda and guiding its execution for local and global impact.

Tackling Barriers to Data Sharing and Registered Reports

Although the benefits of data sharing and registered reports are well-established, researchers often face barriers to adopting these practices. Some of these barriers include fears of data misuse, intellectual property concerns, and competitive pressures in academia (Wicherts et al., 2011). To address these challenges, the network should offer incentives for data sharing, such as grant priority for projects that commit to transparency, and develop robust governance structures that protect intellectual property while promoting data access.

The network should establish a peer-review system specifically for SIL research that evaluates study designs before data collection. This would ensure methodological rigor and reduce biases that often plague SIL studies, such as publication bias and selective reporting.

Conclusion: A Call to Action for Big Team Science in SIL Research

The future of SIL research hinges on our ability to transform how we conduct studies, share data, and collaborate across borders. By adopting Big Team Science principles and developing a dedicated research network, we can tackle the systemic issues identified in the evidence gap maps—small sample sizes, low-quality studies, and a lack of global representation. This will not only improve the quality and generalizability of research but also ensure that interventions are relevant to diverse populations, particularly those in LMICs (see, for a relevant discussion, Akhter-Khan, Ghai, & Mayston, 2025).

Building these large-scale collaborations requires the involvement of all stakeholders—researchers, policymakers, funders, and communities. The time to act is now, and by working together through innovative models like BTS and collaborative networks, we can develop and validate interventions that will truly make a difference across different contexts and populations.