Complex networks articles from across Nature Portfolio

Evaluating socio-technical innovation bundles in participatory action research is challenging because emergent, adaptive and relational changes require multiple forms of evidence and iterative learning. We propose a three-layer framework integrating a stakeholder-owned Theory of Change, periodically revised through a deliberation–mapping–simulation cycle using Fuzzy Cognitive Mapping (cognitive), small-wins tracking with formal indicators (reflective) and Social Network Analysis (relational). Embedded in stakeholder learning cycles, its primary contribution is architectural rather than methodological, bringing distinct forms of evidence into structured dialogue to support complexity-informed evaluation, adaptive management and learning.

Although large language models (LLMs) have only recently become ubiquitous, psychological distress following user-LLM interaction has already been documented in both the scientific literature and the popular media. Such distress may be related to the tendency of some LLMs to provide sycophantic responses that uncritically reinforce the views of the user. To support users’ healthy emotional regulation, flexible, emotionally aware AI systems are needed.

As the term ‘digital twin’ spreads across science and technology, it risks losing scientific meaning. Such a twin should not denote just a digital replica, but a computational model that can be updated, tested against intervention and shown to fail.

The relationship between climate change, poverty, and health is increasingly understood as requiring a complex systems lens in the design of effective and sustainable responses. Climate change exacerbates poverty and health inequities in ways that are non-linear, multidimensional, and deeply interconnected, making traditional intervention approaches insufficient. In this comment, four types of complexity are examined—compositional, experiential, dynamic, and governance—each of which provides a distinct lens for understanding the multilayered interactions at the climate-poverty-health nexus. Building on these dimensions of complexity, complex systems frameworks that provide structured ways to examine and intervene in the climate-poverty-health nexus are explored. Through multiple examples, we illustrate how such approaches can be practically applied. Complex systems-informed analyses can lead to better-designed policies and strategies that are not only more likely to be effective but also more equitable and sustainable in mitigating the impacts of climate change on the health of those living in poverty.

Storytelling is a valuable skill in public health, yet contemporary challenges, such as generative artificial intelligence, are on the rise. Here, we examine some of these challenges and present key strategies for moving storytelling forward as a public health competency.

Neural manifolds refer to the low-dimensional shapes formed by populations of spiking neurons. This intrinsic population geometry can be physically emulated by electronic devices, facilitating near-training-free forecasting with exceptional energy efficiency.