Collective Motion Driven by Social Interactions in Animals

Abstract

There exist two central challenges facing the field of animal movement ecology.
The first of these, addressing the ‘problem of scale’, requires studies which link
the behaviour of individuals to the movement of entire populations. The second,
is to develop a deeper understanding of the feedback mechanisms that exist
between population-level movement and the interactions of individuals, and
also the feedbacks between spatial and social factors that drive movement. To
this end, the aim of this Thesis is to develop a series of mathematical models that
work towards enhancing our understanding of how individual interactions and
behaviour feed into population-level structures and movement. In particular,
we want to explore how social interactions between individuals, and especially
those driven by sexual conflict, can give rise to large scale spatial patterns and
movement in animal populations.
Behaviour driven by sexual conflict is seen across taxa, but one species in
which the behaviour has been widely studied, is Trinidadian Guppies (Poecilia
reticulata). Their habitat and the nature of their interactions gives rise to a natural
separation of spatial, social and temporal scales, making them ideal for studies
in movement ecology. Their behaviour has therefore formed a significant part of the motivation for this Thesis, informing three models of animal movement
across a range of scales.
At each spatial scale, we present a model of animal movement in response
to local social factors such as sex ratio and density and show how a variety
of population-level behaviours can arise from simple assumptions about the
interactions and movement decisions of individuals. At the ‘microscale’, we
briefly explore a coagulation-fragmentation model of group dynamics, and show
that time spent at unfavourable (male-biased) sex ratios can give rise to higher
movement rates of individuals from their patch. At the ‘mesoscale’, we present a
stochastic model of movement in response to sex ratio across a patch network. By
prescribing only simple rules that govern individual responses and movement,
we are able to explore the correlation structure of movements, which give rise
to a wide range of behaviours — from mate searching, to chase and escape
and avoidance behaviour — each revealing unique insights. Finally, at the
‘macroscale’ we consider how density-dependent movement of individuals can
give rise to population-level effects such as dispersal into new territory. We find
that any behavioural mechanism that increases diffusion (movement) at low
densities could have a strong effect on the speed of population dispersal.
By developing mathematical models that span a range of spatial, social and
temporal scales, and also considering how individual behaviours driven by both
social and spatial factors feed into and drive population-level movement, the
work presented in this Thesis will provide a significant contribution towards
addressing key challenges in the field of movement ecology.

Date of Award	22 Apr 2026
Original language	English
Awarding Institution	University of Bath
Supervisor	Richard James (Supervisor) & Tim Rogers (Supervisor)