We discuss self-organized dynamics of agent-based models
with focus on a prototype model driven by non-symmetric self-alignment
introduced in [1].
Unconditional consensus and flocking emerge when the self-alignment is
driven by global interactions with a sufficiently slow decay rate.  In
more realistic models, however, the interaction of self-alignment is
compactly supported, and open questions arise regarding the emergence
of clusters/flocks/consensus, which are related to the propagation of
connectivity of the underlying graph.
In particular, we discuss heterophilious self-alignment: here, the
pairwise interaction between agents increases with the diversity of
their positions and we assert that this diversity enhances
flocking/consensus. The methodology carries over from agent-based to
kinetic and hydrodynamic descriptions.

[1] A new model for self-organized dynamics and its flocking behavior, J.
Stat. Physics 144(5) (2011) 923-947.