Using immersive virtual reality to reveal the dynamical structure of social interactions during collective decision-making

VR environment lab setup

A primary limitation of studying collective behaviour is that reciprocal feedbacks among interacting individuals make it extremely difficult to determine causality (i.e. who influences whom). Even determining what constitutes a social interaction can be very difficult. Consequently, we only have a rudimentary understanding of the nature and structure of communication networks in animal groups. The application of Virtual Reality (VR) environments allows us to experimentally dissociate social input and responses, opening powerful avenues of inquiry into the dynamics of social influence and the physiological and neural mechanisms of collective behaviour. The underlying issue of causality affects almost all research objectives pursued by our centre. Thus, our research will impact the development of future hypotheses and experimental technologies broadly.

CASCB has developed a suite of core facilities that allow researchers to apply virtual environments in experiments with model systems.


Locusts

This project aims to develop concrete tools to systematically characterize how individuals interact with their visual surroundings, with the ultimate goal of better understanding the social interactions and collective decision-making of swarm-forming locust groups. This will be achieved with the LocustVR system, which is the world’s first VR system to study freely walking locusts. Here, unrestrained locusts can be studied as they interact with highly controlled visual projections of surrounding conspecifics moving at different speed, direction, and coherence.


Paul Nuhrenberg, CASCB

Fish

Cichlids

To study social interactions using standardized and manipulable stimuli, this project aims to first record and analyse the behaviour of socially interacting fishes using machine-learning based tracking approaches. The team then creates 3D in silico models of these fishes performing social behaviours, breaking them down into their constituent parts (or behavioural syllables) and presenting them to real fish using a virtual environment. They will observe how the evolutionary radiation that has diversified social phenotypes in Lake Tanganyikan cichlids has led to baseline differences in the elements of social interaction displayed under different social inputs, looking at behaviour as well as patterns of neural activity and hormonal states.

Zebra Fish

Fish are known to adapt their behaviour to physical stimuli (matching swimming speed with the ones from flow tanks), or visual stimuli (optokinetic and optomotor response). Using the immersive VR this project intends to test if zebrafish change their behaviour when placed in a closed-loop immersive VR simulation, where fish will have a perceived displacement different to the one they performed, i.e. the world moves in relation to the fish in a way that they perceive as moving with different velocities. This is the first step in testing how fish can sense and behave if placed in non euclidean environments.