Why We Experience Society Differently: Intrinsic Dispositions as Drivers of Ideological Complexity in Adaptive Social Networks

Understanding the emergence of inequality in complex systems requires attention to both structural dynamics and intrinsic heterogeneity. In the context of opinion dynamics, traditional models relied on static snapshots or assumed homogeneous agent behavior, overlooking how diverse cognitive dispositions shape belief evolution. While some recent models introduce behavioral heterogeneity, they typically focus on macro-level patterns, neglecting the unequal and individualized dynamics that unfold at the agent level. In this study, we analyze an adaptive social network model where each agent exhibits one of three behavioral tendencies-homophily, neophily (attention to novelty), or social conformity-and measure the complexity of individual opinion trajectories using normalized Lempel-Ziv complexity. We find that the dynamics are often counterintuitive-homophilic agents, despite seeking similarity, become increasingly unpredictable; neophilic agents, despite pursuing novelty, exhibit constrained exploration; and conformic agents display a two-phase trajectory, transitioning from early stability to later unpredictability. Moreover, these patterns remain similar across diverse network settings, suggesting that internal behavioral dispositions - rather than external environment alone - play a central role in shaping long-term opinion unpredictability. The broader implication is that individuals’ experiences of ideological volatility, uncertainty, or stability are not merely environmental, but may be endogenously self-structured through their own cognitive tendencies. These results establish a novel individual-level lens on opinion dynamics, where the behavioral identity of agents serves as a dynamical fingerprint in the evolution of belief systems, and gives rise to persistent disparities in dynamical experience within self-organizing social systems, even in structurally similar environments.

💡 Research Summary

This paper investigates how intrinsic behavioral dispositions—homophily (preference for similarity), neophily (attention to novelty), and social conformity—shape the evolution of opinions in adaptive social networks, and how these dispositions generate heterogeneous experiences of ideological volatility at the individual level. Traditional opinion‑dynamics models have largely relied on static snapshots or assumed homogeneous agents, thereby missing the role of cognitive heterogeneity. Recent work introduced behavioral heterogeneity but focused on aggregate outcomes such as overall polarization or network connectivity, obscuring the unequal dynamical trajectories experienced by individual nodes.

The authors adopt the adaptive network framework of Sayama (2018), in which each node i holds a continuous opinion x_i and directed edge weights w_ij encode the flow of information from j to i. Opinion dynamics follow
dx_i/dt = c (⟨x⟩_i – x_i) + ε,
where ⟨x⟩_i is the weighted average of neighbors’ opinions (the perceived social norm) and c is the conformity strength. Edge weights evolve according to two behavioral functions:
dw_ij/dt = h F_h(x_i, x_j) + a F_a(⟨x⟩_i, x_j).
F_h = θ_h – |x_i – x_j| captures homophilic reinforcement (similar opinions strengthen ties), while F_a = |⟨x⟩_i – x_j| – θ_a captures neophilic reinforcement (ties strengthen when a neighbor’s opinion deviates from the local norm). Parameters h, a, and c are drawn from distributions that encode each node’s dominant disposition.

To quantify the unpredictability of each node’s opinion trajectory, the study uses normalized Lempel‑Ziv (nLZ) complexity. Continuous opinion values are discretized into symbolic bins (Δ = 0.75) and the LZ parsing algorithm counts the number of distinct substrings. The raw LZ count is normalized by n log n (where n is the sequence length) to obtain nLZ, which is independent of trajectory length. High nLZ indicates low compressibility and thus high temporal unpredictability.

Three experimental scenarios are explored with networks of 300 nodes simulated for 3,000 time steps, repeated ten times per configuration:

1. Pure networks – all nodes share the same disposition (homophilic, neophilic, or conformist).
2. Dual‑faction networks – two equally sized factions with opposing dispositions (e.g., homophily vs. neophily).
3. Mixed networks – each node’s parameters are drawn uniformly, producing a fully heterogeneous population.

For each run, nLZ is computed over expanding windows (t = 0 to 3000 in steps of 300) and averaged across nodes of the same type. Bootstrap resampling yields 95 % confidence bands.

Key Findings

• Homophilic agents display a monotonic increase in nLZ. Although they seek similarity, the continual reinforcement of like‑minded ties creates a cascade of small opinion mismatches and evolving neighborhood structures, which compound over time and make their opinion paths increasingly unpredictable. This counter‑intuitive result suggests that strong homophily can be a source of long‑term instability.

• Neophilic agents quickly settle into a low, stable nLZ plateau. Their drive to connect with novel opinions leads to early exposure to diverse viewpoints, but the adaptive rewiring quickly channels these interactions into a bounded set of symbolic patterns. Consequently, despite continual novelty seeking, their trajectories remain relatively predictable, indicating that novelty can paradoxically constrain exploration when the network self‑organizes around it.

• Conformist agents exhibit a biphasic trajectory. In early stages, conformity to the local norm suppresses variability, causing nLZ to decline or remain flat. After a critical period, the accumulated structural changes and influx of divergent opinions trigger a rapid rise in nLZ, reflecting a sudden loss of predictability. This two‑phase pattern highlights conformity’s role in providing short‑term stability that can give way to abrupt volatility.

Importantly, these patterns persist across all three scenarios. Whether the population is homogeneous, split into competing factions, or fully heterogeneous, each behavioral class retains its characteristic nLZ evolution. Thus, the internal disposition of agents, rather than the external network topology, is the dominant determinant of long‑term opinion unpredictability.

Theoretical Contributions

1. Introduces a fine‑grained, node‑level metric (nLZ) to capture temporal complexity of opinion dynamics, moving beyond static or short‑window measures.
2. Demonstrates that behavioral heterogeneity produces persistent disparities in “ideological experience” – some individuals live in a world of escalating volatility, others in a relatively stable environment, even when embedded in the same structural context.
3. Provides evidence that intrinsic cognitive tendencies can self‑organize the macro‑level pattern of polarization and fragmentation, offering a mechanistic bridge between micro‑level psychology and macro‑level social inequality.

Implications and Future Directions

The findings suggest that interventions aimed at reducing polarization should consider not only network rewiring (e.g., promoting cross‑cutting ties) but also the distribution of behavioral dispositions within the population. Moreover, the nLZ framework could be applied to empirical data from online platforms to identify users at risk of experiencing high ideological volatility. Extending the model to incorporate additional cognitive biases (confirmation bias, affective contagion) or external shocks (media events) would further enrich our understanding of how personal dispositions and exogenous forces jointly shape the dynamical landscape of public opinion.