https://arxiv.org/api/2WQXeVim1yo1U/qC0bZrhjKyFNg 2026-09-11T20:41:35Z 8341 30 15 http://arxiv.org/abs/2606.03071v2 2026-08-29T19:50:04Z

We consider continuous Prisoner's Dilemma games played in a spatial setting by group-structured populations. The evolutionary dynamics is driven by within-group individual-level birth and death events, and by group-level fission and extinction events. Within groups, individuals play well-mixed games, while groups play games on a 1-dimensional spatial grid against their nearest neighbours. Payoffs from individual-level games affect birth rates of individuals, and payoffs from group-level games affect group extinction and fission probabilities. It is well known that the within-group evolution by itself always results in a complete loss of cooperation, and that in the group-level games, defection is also favoured if these games are played in well-mixed populations of groups. Here we show that despite this double disadvantage, cooperation can be maintained due to spatially structured group-level dynamics. Mirroring results from one-level selection models, the spatial nature of games between groups and the resulting fissioning and extinction events is essential for the evolution of cooperation in these models, but needs to manifest itself in specific ways in order to be effective: we find that higher levels of cooperation evolve when the selection generated by games between groups acts locally rather than globally.

2026-06-02T03:00:53Z 25 pages 7 figures Yaroslav Ispolatov Burton Simon Michael Doebeli http://arxiv.org/abs/2608.29174v1 2026-08-29T09:52:00Z

Large language models (LLMs) are increasingly adopted as closed-loop controllers in physical multi-agent systems, yet their emergent collective dynamics remain incompletely characterised. We deploy 22 LLM agents as direct, real-time target-speed controllers (per 0.5 s cycle, with IDM as collision-avoidance clamp) on a 230 m ring road under the Sugiyama 2008 paradigm, reproducing human-like stop-and-go waves. Six matched controls spanning stochasticity (white noise, OU noise, temperature), population variance, and dynamical instability (delay, OV model) are systematically excluded. The surviving phenomenon, termed Sustained Heterogeneity (SH), is the persistent, approximately temperature-insensitive (approx. 8 percent across a 6x T sweep), per-cycle divergence in LLM-chosen target-speed adjustments, propagating through a three-stage cascade of drift, gap erosion, and nonlinear braking. Across four traffic densities, the critical LLM penetration fraction p_c decreases monotonically from no transition at density 43.5 veh/km to p_c approx 0.23 at density 95.7 veh/km, consistent with an initiation-threshold model governed by trigger distance, stochasticity, and fleet size. Chain-of-thought analysis of 39,600 decisions across three seeds shows agents engage in multi-factor safety reasoning, yet systematic divergence persists, implying stability must be enforced at the dynamics layer. This is the first study to identify a previously uncharacterised collective mechanism in LLM-controlled traffic and map a density-dependent phase boundary p_c(rho).

2026-08-29T09:52:00Z 41 pages, 5 figures Yujun Qi Yangyang Guan http://arxiv.org/abs/2608.29166v1 2026-08-29T09:31:23Z

Synchronization is conventionally understood as the emergence of a phase-locked collective state among interacting dynamical units. However, extending this notion to systems whose dynamical variables are defined on higher-dimensional simplices introduces fundamentally new constraints arising from the topology of the underlying simplicial complex. In the simplicial Kuramoto model, nontrivial topological cycles give rise to a higher dimensional harmonic subspace that is unaffected by the coupling and can therefore drift indefinitely, preventing a globally synchronized state. To resolve this we employ Hodge decomposition on simplicial Kuramoto dynamics and investigated the components. Despite this topological obstruction, we show that the simplicial Kuramoto model admits fixed-point states in the exact and coexact sectors of the Hodge decomposition above critical coupling strengths, which we derive analytically for both sectors. This decomposition provides a generalized notion of synchronization in which the non-harmonic components converge to fixed states. We further investigate the robustness of these fixed-point states to external perturbations. Excluding the drifting and non-interacting harmonic component, the perturbation response is governed by the spectrum of weighted Laplacian and recovers a Kirchhoff index dependence analogous to standard Kuramoto case. In contrast, the harmonic sector is non-dissipative, and consequently the fragility of the system increases with the dimension of this topological subspace. We characterize this effect numerically using triangulated tori with varying first Betti number and find a superlinear scaling relation between system fragility and the dimension of the harmonic sector.

2026-08-29T09:31:23Z 14 pages, 11 figures Abhijeet Kumar Palash Kumar Pal Dibakar Ghosh http://arxiv.org/abs/2608.28320v1 2026-08-28T13:31:33Z

Anticipating the onset of collective synchronization is important in many networked systems, yet observing every oscillator is often impractical. We investigate whether synchronization transitions can be detected from a small set of monitored, or sentinel, nodes. Using a stochastic Kuramoto model on networks, we numerically compare three early warning signals: the local order parameter, its temporal variance, and the variance of individual oscillator phases after removing their mean rotational trends. We also compare sentinel-selection strategies based on node dynamics, degree, and random sampling. We show that, under partial observation, the local order parameter and the variance of detrended phases provide substantially stronger warning signals than the variance of the local order parameter. Selecting nodes according to their dynamical behavior near the transition consistently improves performance over other sentinel-selection methods. With only $\lfloor\ln N\rfloor$ dynamically selected sentinels, two success indicators approach the performance obtained by observing all $N$ nodes. These results demonstrate that synchronization transitions can be anticipated from sparse observations when the warning signal and monitored nodes are chosen appropriately.

2026-08-28T13:31:33Z 14 pages, 5 figures, 3 tables Yusuke Kato Naoki Masuda http://arxiv.org/abs/2608.28046v1 2026-08-28T08:08:39Z

Collective behaviour in living systems is usually modelled as the outcome of a \emph{direct} social drive: agents are rewarded, or hard-wired, to align with or approach their neighbours. Here we show that aggregation can instead emerge from an \emph{indirect} objective. We let reinforcement learning foragers, initially performing a random walk, optimize their dynamics from a purely individual reward for finding replenishable targets, while perceiving only their conspecifics and never the targets themselves. As the visual range grows, the agents undergo a sharp crossover from an environment-tuned individual search to a scale-agnostic collective one, and this crossover coincides with the onset of spatial aggregation. Thus a collective phase arises as a by-product of optimal foraging, without any direct reward for grouping. A minimal analytical first-passage model reproduces the transition as a crossover between the two search strategies. Our results identify indirect, resource-driven reward as a generic route to emergent collective phenomena.

2026-08-28T08:08:39Z 11 pages, 8 figures Gorka Muñoz-Gil Andrea López-Incera Vide Ramsten Giovanni Volpe Thomas Müller Hans J. Briegel http://arxiv.org/abs/2608.27751v1 2026-08-27T22:29:05Z

How do organisms navigate when the signals guiding them are noisy? Variance adaptation, the rescaling of sensitivity to noise, is common in sensory systems, but its role in navigation is unexplored. We introduce a minimal active Brownian particle whose run-and-tumble policy follows from an optimality principle. Variance adaptation emerges as part of this policy. Adaptation keeps chemotactic drift finite as noise grows, while a non-adaptive particle's collapses exponentially. Adaptation also carries a cost, degrading performance in quiet environments and requiring a tuned adaptation sensitivity.

2026-08-27T22:29:05Z 5 pages, 4 figures, 4 supplemental files Aniruddha Datta Shiladitya Banerjee http://arxiv.org/abs/2608.27436v1 2026-08-27T17:54:34Z

Low-dimensional reduction theories such as the Ott-Antonsen ansatz have played a crucial role in the study of populations of coupled oscillators. Their application, however, has largely been restricted to systems with frequency distributions of rational-function form, such as the Cauchy distribution. For such distributions, the residue theorem allows the dynamics of the global order parameters to be closed in terms of a finite number of poles, thereby yielding a finite-dimensional system of ordinary differential equations. Rational frequency distributions, however, generally have heavy tails and only finitely many well-defined moments, and therefore may not always be realistic as frequency distributions. In this paper, we develop an approximate low-dimensional reduction theory based on perturbation theory for weakly heterogeneous populations of phase oscillators with a Gaussian frequency distribution. We construct the theory not only for first-harmonic coupling, for which the Ott-Antonsen ansatz applies, but also for populations of phase oscillators with multi-harmonic coupling. The effectiveness of the proposed low-dimensional reductions is demonstrated through both theoretical analysis and numerical simulations.

2026-08-27T17:54:34Z 11 pages, 3 figures, Kai Tokunaga http://arxiv.org/abs/2608.23993v2 2026-08-27T08:11:01Z

In oscillator networks sustained by finite resources, synchronization can depend on both the total resource and its spatial distribution. We study a duplex system whose activity layer consists of chaotic Hindmarsh--Rose oscillators and whose transport layer redistributes a conserved resource through a degree-biased Markov process. The stationary resource field is characterized analytically, and its existence, uniqueness, and convergence are established. By embedding this field into a local adaptive feedback law, the available resource is converted into node-dependent dissipation, for which Lyapunov analysis guarantees convergence to the synchronization manifold. Numerical results show that topology bias reorganizes the transient route to synchronization. Weak bias produces an almost collective contraction, whereas intermediate bias creates a hub-initiated recruitment hierarchy that extends toward middle-degree and peripheral nodes. Under stronger bias, the degree hierarchy becomes more pronounced while peripheral recruitment slows because adaptive dissipation is concentrated on structurally privileged nodes. Across the explored parameter range, this localization--coverage tradeoff is accompanied by a non-monotonic synchronization response at fixed total resource. The largest Lyapunov exponent remains positive after synchronization and the correlation dimension changes only modestly, consistent with suppression of transverse deviations while chaotic motion is retained on the synchronization manifold.

2026-08-25T02:33:58Z Zhouqi Li Xiaoyan He Yuanhong Bi Zengping Zhang http://arxiv.org/abs/2608.26323v1 2026-08-26T18:55:07Z

Mixed-mode oscillations (MMOs), characterized by the alternation of small-amplitude oscillations (SAOs) and large-amplitude oscillations (LAOs), and bursting oscillations are common forms of complex oscillatory dynamics observed in systems with multiple timescales and have been widely studied across scientific disciplines. Mixed-mode bursting oscillations (MMBOs) combine features of MMOs and bursting, with LAOs organized into burst events. Most existing studies treat MMBOs as two-timescale phenomena, identifying distinct geometric mechanisms depending on how the timescales are grouped. In this work, we use a three-timescale biophysical cortical neuronal oscillator model to demonstrate that a three-timescale implementation of geometric singular perturbation theory (GSPT) provides stronger predictive insight into MMBO dynamics. This perspective unifies mechanisms previously identified from fast-slow analysis, while revealing the canard-delayed-Hopf (CDH) singularity as an organizing center for MMBOs near the singular limit. We perform a detailed bifurcation analysis of the full eight-dimensional model to determine how MMBOs are organized along families of isolas. We then combine GSPT with full-system bifurcation analysis to show how tuning the relative timescales induces transitions among MMOs, MMBOs, and bursting dynamics. Our results highlight the importance of studying MMBOs from a three-timescale perspective and demonstrate that combining GSPT with full-system bifurcation analysis can reveal organizing structures and mechanisms that are not apparent from either approach alone.

2026-08-26T18:55:07Z Ngoc Anh Phan Yangyang Wang http://arxiv.org/abs/2608.25995v1 2026-08-26T16:46:44Z

In the Derrida--Higgs model of sympatric speciation, a sexually reproducing population with finite binary genomes can mate only when the genetic overlap between two individuals exceeds a threshold $q_{\min}$. Depending on the genome length $L$, the population either remains genetically connected or fragments into reproductively isolated species. A central problem is therefore to predict the critical genome length $L_c$ at which fragmentation begins. At finite $L$, fluctuations broaden the overlap distribution and allow genetically distant regions of the population to remain connected. A previously proposed transient variance criterion captures this effect, but its evaluation requires numerical iteration of coupled moment equations. Here we first correct the unrestricted moment equations by removing a previously implicit assumption and then derive an explicit closed form expression for $L_c$. The resulting formula shows that the critical genome length is determined, at the time the mean overlap reaches $q_{\min}$, by the competition between deterministic separation from the unrestricted equilibrium and the transient genealogical variance of the overlap distribution. This expression permits a systematic asymptotic analysis. When the deterministic contribution dominates, $L_c$ becomes essentially independent of the population size $M$ and scales as $μ^{-2}$. When the transient genealogical variance dominates, $L_c$ grows as $M^{3/2}$ or as $\sqrt{M}/μ$, depending on how $M$ and the mutation rate $μ$ jointly vary. Simulations support all predicted behaviors. Our results identify transient genealogical variance as the principal mechanism linking finite genome fluctuations to the onset of reproductive fragmentation and provide a practical analytical prediction for the critical genome length.

2026-08-26T16:46:44Z 56 pages, 9 figures, + appendices Dan Braha Marcus A. M. de Aguiar Vitor M. Marquioni http://arxiv.org/abs/2506.14006v2 2026-08-25T18:37:47Z

We present a framework for chemical learning based on Competitive Dimerization Networks (CDNs) - systems in which multiple molecular species, e.g., proteins, DNA oligomers, or RNA oligomers, reversibly bind to form dimers. We show numerically that these networks can, in principle, be trained in vitro through directed evolution, enabling the implementation of complex learning tasks such as multiclass classification without digital hardware or prior knowledge of all microscopic association constants. Each molecular species functions analogously to a neuron, with binding affinities acting as tunable synaptic weights. A training protocol involving mutation, selection, and amplification of DNA-based components allows CDNs to robustly discriminate among noisy input patterns. The resulting classifiers exhibit strong output contrast and high mutual information between input and output, especially when guided by a contrast-enhancing loss function. Comparative analysis with in silico gradient descent training reveals closely correlated performance. These results establish CDNs as a promising platform for analog physical computation, bridging synthetic biology and machine learning, and advancing the development of adaptive, energy-efficient molecular computing systems.

2025-06-16T21:10:12Z 10 pages, 5 figures + Appendix Alexei V. Tkachenko Bortolo Matteo Mognetti Sergei Maslov http://arxiv.org/abs/2607.10381v2 2026-08-25T18:20:59Z

We investigate the non-equilibrium phase transitions of the continuous-trait Axelrod model, an agent-based framework where individual culture is represented by a vector of $F$ continuous features confined to the interval $(0,1)$. Local interactions are governed by a metric similarity threshold $d$, which acts as a continuous control parameter of social tolerance. The dynamics inevitably freeze into one of two absorbing configuration classes: an ordered, homogeneous monocultural state at high tolerance, or a highly fragmented, disordered state at low tolerance. While previous studies characterized the transition as hybrid based on the continuous behavior of the domain density $μ$ alongside a discontinuous jump in the largest domain fraction $ρ$, we show that this apparent continuity is an artifact of severe finite-size masking effects. By shifting the methodological focus to the scaling of the median $\tildeμ$ and analyzing the full probability distributions $P(μ)$, we unveil a clear bimodal structure with disjoint maxima across independent simulation runs. Our results reveal that for $F=2$, the system undergoes a genuinely hybrid transition in the contemporary sense, featuring a tiny but finite latent jump ($μ_c \approx 0.089$) at the critical threshold $d_c \approx 0.0784$ while scaling toward it from below via a non-analytical power law with a mean-field exponent $β\approx 1/2$. Conversely, for $F=3$, the higher trait-space dimensionality suppresses local fluctuations, yielding a traditional, non-hybrid first-order transition. We apply this framework to the alternative discrete Poisson variant of the model, successfully confirming its known continuous transition for $F=2$ and discontinuous, non-hybrid transition for $F=3$, thereby establishing a unified characterization of phase transitions in Axelrod-like systems.

2026-07-11T16:17:54Z Sandro M. Reia Paulo R. A. Campos José F. Fontanari http://arxiv.org/abs/2512.14729v2 2026-08-24T21:40:48Z

Topological higher-order synchronization reveals collective phenomena demonstrating how topology shapes dynamics, yet providing a general theoretical framework for designing and controlling dynamical states on higher-order networks remains a challenge of the field. Here we propose a dynamical systems framework for topological cluster synchronization on directed hypergraphs that can be used to design synchronization patterns on nodes and hyperedges of directed hypergraphs. By encoding oriented hyperedges into the hypergraph topological Dirac Hamiltonian, we obtain a spectrum whose isolated eigenstates correspond to distinct topological synchronization cluster states defined jointly on nodes and hyperedges. By selecting any isolated eigenstate, the system can be driven toward the associated dynamical state reflecting a specific partition of the hypergraph without modifying the underlying hypergraph structure. We numerically demonstrate the ability to design different topological cluster synchronization states on directed-hypergraph block models and empirical systems-including higher-order contact networks and the ABIDE functional brain network. Our results establish a general and interpretable route for controlling collective dynamics in directed higher-order systems.

2025-12-10T04:32:40Z (22 pages, 9 figures) Yupeng Guo Ahmed A. A. Zaid Xueming Liu Ginestra Bianconi http://arxiv.org/abs/2510.09387v2 2026-08-24T17:26:17Z

Understanding how higher-order interactions affect collective behavior is a central problem in nonlinear dynamics and complex systems. Most works have focused on a single higher-order coupling function, neglecting other viable choices. Here we study coupled oscillators with dyadic and three different types of higher-order couplings. By analyzing the stability of different twisted states on rings, we show that many states are stable only for certain combinations of higher-order couplings, and thus the full range of system dynamics cannot be observed unless all types of higher-order couplings are simultaneously considered.

2025-10-10T13:41:40Z Per Sebastian Skardal Federico Battiston Maxime Lucas Matthew S Mizuhara Giovanni Petri Yuanzhao Zhang http://arxiv.org/abs/2608.23426v1 2026-08-24T16:05:27Z

The Kuramoto model provides a paradigmatic framework for studying synchronization of interacting oscillators, and has been generalized to arbitrary dimensions to describe swarms, flocks and multi-dimensional opinion dynamics. Yet, existing formulations neglect inertia, a key mechanism known to enhance information propagation and collective responsiveness. Here, we introduce and analyze an inertial Kuramoto model in arbitrary dimensions. We show that inertia fundamentally alters the nature of the synchronization transition, inducing a crossover from continuous to discontinuous behavior, with the onset of hysteresis depending explicitly on both inertia and dimensionality parity. Our analytical theory is supported by extensive numerical simulations. These results establish inertia as a crucial ingredient of high-dimensional collective dynamics and reveal a novel universal structure in synchronization phenomena.

2026-08-24T16:05:27Z 7 pages, 4 figures, accepted in Phys. Rev. E Kirill Kovalenko Bruce X. Dai Fanshu Fang Zerong Guo Haoran Liu Federico Botta Charo I. del Genio Stefano Boccaletti Simona Olmi