--- title: Dynamic Social Behavior Model (DSBM) url: https://www.emergentmind.com/topics/dynamic-social-behavior-model-dsbm type: topic --- # Dynamic Social Behavior Model (DSBM) Searching arXiv for the cited DSBM-related papers and neighboring terminology. Searching arXiv for “1512.07607 dynamic social networks based on movement”, “1403.0921 dynamic stochastic blockmodels”, and related DSBM terminology. Dynamic Social Behavior Model (DSBM) denotes a class of dynamic models in which latent social structure and observed behavior are separated and linked over time. In the usage most directly aligned with the term, it is a Bayesian hierarchical, discrete-time, continuous-space hidden Markov model in which an unobserved, time-varying social network controls movement through attraction and alignment, allowing social relationships to be inferred from telemetry rather than from direct behavioral observation [1512.07607]. In adjacent network science, the same acronym also appears as “Dynamic Stochastic Blockmodel,” a distinct state-space model for evolving block interaction structure in social networks [1403.0921]. ## 1. Scope and defining idea In the movement-based formulation, the DSBM is designed for settings in which direct behavioral observations are expensive, opportunistic, invasive, or unavailable, but repeated movement measurements are available. The central hypothesis is that sociality leaves a movement signature: socially connected individuals tend to move toward one another and to travel in similar directions. The model therefore separates three layers: the **latent social behavior/state**, the **behavioral expression** through movement, and the **observation process** through noisy telemetry [1512.07607]. This separation is methodologically important. Trajectories are not treated as the social network itself; they are treated as noisy consequences of an unobserved dynamic network. A plausible implication is that the DSBM is best understood as a latent-state model of social affiliation rather than as a proximity heuristic or a pure movement model. The distinction matters because the model is intended to recover who is socially connected to whom, when ties form or dissolve, and how those ties persist, rather than merely to threshold distances between individuals. ## 2. Latent dynamic social network The core latent object is a dynamic network \(\mathbf W(t)\) indexed at discrete times \(t=1,\dots,T\), with binary, undirected, unweighted edges \[ w_{ij}(t)= \begin{cases} 1, & \text{if individuals } i \text{ and } j \text{ are socially connected at time } t,\\ 0, & \text{otherwise.} \end{cases} \] Each edge evolves as a pairwise independent Bernoulli Markov chain. At the initial time, \[ w_{ij}(1)\sim \mathrm{Bern}(p_1), \] and for \(t=2,\dots,T\), \[ w_{ij}(t)\mid w_{ij}(t-1)\sim \begin{cases} \mathrm{Bern}(p_{1|0}), & w_{ij}(t-1)=0,\\ \mathrm{Bern}(p_{1|1}), & w_{ij}(t-1)=1. \end{cases} \] Here \(p_1\) is the marginal edge probability at \(t=1\), \(p_{1|0}\) is the formation probability, and \(p_{1|1}\) is the persistence probability [1512.07607]. The model assumes that expected network density is constant in time and reparameterizes the transition law through a temporal-stability parameter \(\phi\): \[ p_{1|0}\equiv (1-\phi)p_1, \qquad p_{1|1}=1-(1-\phi)(1-p_1). \] This yields a particularly transparent interpretation: \(\phi=0\) corresponds to temporal independence, because then \(p_{1|0}=p_{1|1}=p_1\), while \(\phi\to 1\) corresponds to a near-static network in which edge states almost never change. Within this construction, the DSBM encodes social behavior as a time-varying binary relation with explicit formation and persistence dynamics rather than as a static latent partition or a distance threshold. ## 3. Movement process: attraction and alignment Let \(\mu_i(t)\in\mathbb R^2\) denote the latent position of individual \(i\) at synchronized time \(t\). Conditional on the social network, the positions evolve according to a Gaussian Markov random field whose mean and precision depend on current ties. The model uses two mechanisms. The first is **attraction**. For each individual, the centroid of connected neighbors is \[ \bar\mu_i(t)\equiv \sum_{j\neq i}^n \frac{w_{ij}(t)}{w^c_{i+}(t)}\mu_j(t), \] where \(w^c_{i+}(t)\) is the size of the ego-network if the individual has at least one connection, and equals a positive constant \(c\) otherwise. The attraction direction is the unit vector \[ \tilde\mu_i(t)\equiv \begin{cases} \dfrac{\bar\mu_i(t)-\mu_i(t)}{\|\bar\mu_i(t)-\mu_i(t)\|_2}, & \sum_{j\neq i} w_{ij}(t)>0,\\[1ex] 0, & \sum_{j\neq i} w_{ij}(t)=0. \end{cases} \] The use of a unit vector is deliberate: attraction specifies direction while keeping step magnitude approximately bounded. On average, attraction moves an individual a distance \(\beta\) toward its ego-network centroid per time step [1512.07607]. The second mechanism is **alignment**. Its contribution to the conditional mean is \[ \sum_{j\neq i}\alpha \frac{w_{ij}(t)}{w^c_{i+}(t)} \left[\mu_j(t)-(\mu_j(t-1)+\beta \tilde\mu_j(t-1))\right]. \] Only currently connected individuals contribute; the term averages over connected neighbors, subtracts off the attraction component of each neighbor’s displacement, and then aligns residual motion directions. The authors interpret \(\alpha=0\) as no alignment and \(\alpha\to 1\) as near-perfect alignment, while excluding \(\alpha=1\) because it yields an intrinsic conditional autoregressive model with improper covariance. In the joint GMRF form, the precision blocks are \[ Q_{ij}(t)\equiv \begin{cases} -\alpha w_{ij}(t)\sigma^{-2}\mathbf I_2, & j\neq i,\\ w^c_{i+}(t)\sigma^{-2}\mathbf I_2, & j=i. \end{cases} \] These two mechanisms provide the model’s explicit social-behavior semantics: connected individuals tend both to move toward one another and to travel in parallel. ## 4. Observation model and inference The latent positions are not observed directly. Instead, telemetry yields noisy observations \(s_i(\tau_i)\) at irregular, asynchronous times, with individual-specific schedules and measurement error. Direct joint inference over the full posterior is described as computationally infeasible because of irregular observation times, measurement error, asynchronous sampling, and high-dimensional latent trajectory vectors. The paper therefore uses multiple imputation based on a separate continuous-time movement model [1512.07607]. The approximation is \[ [\theta\mid s]\approx \int [\theta\mid \mu=\mu^*]\,[\mu^*\mid s]\,d\mu^*, \] with imputed paths \([\mu^*\mid s]\) generated by the continuous-time correlated random walk model of Johnson et al. (2008), implemented in the R package `crawl`. That imputation model uses an Ornstein–Uhlenbeck process for velocity, treats individuals as conditionally independent at the imputation stage, and accounts for measurement error and irregular observation times. The resulting two-step procedure draws \(K\) imputed paths and, during MCMC for the social model, conditions each iteration on one imputed path; in the application discussed in the paper, \(K\approx 50\) was sufficient. Inference for the social model itself is performed by MCMC. Because \(\mathbf W\) is latent and binary for each pair and time, the number of latent network variables is \[ {n\choose 2}T. \] The authors identify the number of individuals \(n\) as the primary computational bottleneck and report that, once \(n\) exceeds a few dozen, MCMC on a laptop becomes infeasible. They suggest two extensions: imposing a maximum interaction radius \(R_{\max}\), so distant edges are deterministically zero, and imposing stronger structure on \(\mathbf W\), such as clique or subgroup structure, to reduce dimensionality. A further inferential difficulty is the temporal-stability parameter \(\phi\), which is described as hard to estimate and prior-sensitive; in simulation it shows moderate systematic bias toward zero, although posterior inference for \(\mathbf W\) is reported as robust. ## 5. Empirical behavior and validation In simulation, the model is fit to latent paths \(\mu\) directly, so the study isolates recovery of dynamic social structure rather than errors induced by telemetry preprocessing. Performance is compared to a naive proximity network \[ W^R_{ij}(t)=I_{\|\mu_i-\mu_j\|_2