Our research group uses quantitative approaches to study the evolution and adaptive value of animal behaviour in natural contexts. We are interested in how behaviour has evolved, what the adaptive value of behaviour is, and what the mechanisms that underlie behaviour are.
Although we are traditional behavioural ecologists at heart, we borrow computational approaches developed for model laboratory systems like Drosophila and Zebrafish, and employ them in settings where animal behaviour has evolved – Lake Tanganyika, the Mediterranean Sea, coral reefs, and tropical rainforests. Using techniques like computer vision and machine learning, automated tracking of behaviour, and 3D reconstruction of environments, we aim for a quantitative assessment of the expression and value of behaviour in the places it naturally occurs.
Using many different taxa, we seek to understand how social interactions are modified by current context, how animals perceive and process social cues, and how environments – both social and physical – change and are changed by behaviour. We take a broad approach, combining studies of proximate neurobiological and genetic mechanisms, with analyses of the physics of interactions, up to broad evolutionary and ecological studies of social influence and behaviour.
Here are some of our current projects.
The evolution of social behaviour
When complex social structures evolve, what needs to change in terms of the behaviour animals express? Do social animals need to do more, that is, must a richer or more complex behavioural repertoire evolve? Are we as human observers able to detect the potentially subtle ways that behaviour may differ in what appear to be similar contexts? And can machine learning approaches help us in this endeavour?
We ask these questions across a range of systems, from the explosive adaptive radiation in Tanganyikan cichlids, damselfish on Jamaican coral reefs, Trinidadian guppies, and spiders in the Latin American rainforests.
The interaction between physical and social spaces
“We shape our buildings; thereafter they shape us” – Winston Churchill
Animals affect, and are affected by, their environments. This simple relationship means that it is difficult to assess how removing animals from their natural contexts might influence their behavioural expression. We aim to quantify the natural structures in which animals live and interact, and manipulate attributes of these structures to experimentally test their effects.
Sexual selection, mate choice, and parental care in complex social conditions
Payoffs of reproductive strategies vary depending on the social context in which they are employed, but quantifying how social conditions change can be difficult. With spiders and fish, we manipulate elements of the social context, track all interactions among individuals, and measure the fitness payoffs of varying strategies.
The evolution of social competence and cognition
Living in groups potentially imposes cognitive challenges that solitary animals do not face. Memory of past interactions, inference of unknown relationships, and recognition of individuals may be required to maintain functional social groups. Using a blend of virtual and traditional approaches, we try to understand the range of cognitive abilities that have evolved along with social living, along with how these differences may be encoded in the brain itself.