Human Dynamics in Temporal Networks

Human dynamics in temporal networks examines how patterns of individual and collective interactions evolve over time and influence the behaviour of social, technological and biological systems. Unlike static networks, temporal networks capture the sequence and timing of contacts, revealing bursty activity characterised by rapid sequences of events separated by long intervals of inactivity. Such dynamics underlie information diffusion, epidemic spreading and coordination processes. Recent advances have extended models beyond pairwise contacts to incorporate higher-order structures, recognising that real-world interactions frequently occur in groups rather than dyads. Concurrently, non-Markovian processes—where the probability of an event depends on its history—have been integrated into theoretical frameworks, enabling more accurate predictions of transient and steady-state behaviours. Analytic and computational methods now allow the generation of synthetic temporal networks with prescribed inter-event distributions, the characterisation of temporal-topological correlations across interaction orders and the assessment of resilience or vulnerability in spreading phenomena. Together, these developments offer a unified perspective on how timing, memory and group structure shape the emergent properties of dynamic human systems and inform the design of interventions in public health, communication platforms and organisational management.

Research from Nature Portfolio

Recent studies have developed systematic approaches to construct temporal networks that faithfully reproduce empirical burstiness by specifying target inter-event time distributions for nodes and links regardless of underlying topology. Complementary work has introduced higher-order, non-Markovian contagion models in simplicial complexes, revealing that group interactions and memory effects jointly determine thresholds and resilience in spreading processes. Further investigations have characterised temporal-topological patterns of higher-order events across diverse real-world systems, demonstrating that events of different orders cluster both in time and topology and that individuals maintain consistent activity levels across interaction orders. These contributions collectively enhance the capacity to simulate and analyse dynamic processes on networks with realistic temporal correlations and multi-node interactions, bridging theoretical developments with empirically grounded observations.

Research from all publishers

An analysis of information transmission within a mobile drone swarm has applied temporal-network methodology to quantify message delivery times as a function of swarm mobility and communication range, extending traditional graph algorithms to time-varying links and highlighting design principles for efficient autonomous systems. In studies of online forum behaviour, researchers have shown that the heavy-tailed distributions of post clicks and replies are interdependent, suggesting that skewed engagement metrics at the micro level give rise to collective burstiness in user interactions. Investigations of high-activity group messaging platforms have uncovered transitions between bimodal, double-power-law and single-power-law inter-event time distributions, attributing these shifts to the interplay of circadian rhythms and intense collective engagement. Together, these works illustrate the broad applicability of temporal-network concepts—from engineered swarms to digital social systems—and underscore the importance of both endogenous rhythms and network structure in shaping human dynamic patterns.

Human Dynamics in Temporal Networks publication trend

The graph below shows the total number of articles in human dynamics in temporal networks across all publications each year (not limited to Nature Index journals).

Technical terms

Temporal network: A representation of a system in which nodes and edges appear or disappear over time, capturing the order and timing of interactions.

Inter-event time: The elapsed time between consecutive interactions or events involving the same node or link.

Burstiness: A characteristic of activity patterns marked by clusters of rapid events separated by long inactive periods, often yielding heavy-tailed inter-event time distributions.

Non-Markovian process: A stochastic process in which the probability of future events depends on the history of past events, rather than solely on the current state.

Simplicial complex: A mathematical structure generalising networks to include higher-order interactions among three or more nodes, such as triangles or tetrahedra.

References

1. Constructing temporal networks with bursty activity patterns. Nature Communications (2023).
2. Information Transmission in a Drone Swarm: A Temporal Network Analysis. Drones (2024).
3. Higher-order non-Markovian social contagions in simplicial complexes. Communications Physics (2024).
4. Skewed distribution and the heterogeneity of collective clicking and replying behaviours. Systems Science & Control Engineering (2024).
5. Temporal-topological properties of higher-order evolving networks. Scientific Reports (2023).
6. Transition between distribution patterns in human dynamics with high activity. Physical Review Research (2023).