Physics of Collective Animal Behaviour: Making a leap from inanimate particles and atoms to conscious social animals.

– Aawaz Pokhrel and Pranav Kayastha

A Day in the Life of Puckett Lab:

I think it is rather difficult to write about a typical day in our lab, mainly because everyday is at least somewhat different from the other. I guess it is the same smell of fish everyday, and the same flavour of coffee every morning, but what we do here is less predictable and more open ended. One reason why our days are not well scheduled and planned is because not a lot people have looked into collective animal behaviour from the lens of physics, and no one has done projects like that of ours. This means that we will not know what stumbling blocks are patiently waiting for us and that we will have to (try to) solve them on our own. I like to think that we are like an explorer venturing her way into the unknown and the uncharted, even though she knows the journey is going to be frustrating and tedious, because she is confident that when she gets there, her journey will shed light to her fellow travellers and the rest of mankind.


Figure 1. Collective behaviour in different species.

Throughout the summer so far, we have been facing multiple hurdles on our journey to gather and analyse data. Most of them are very open-ended with more than one solutions. How to keep the water temperature constant but without inducing a current? How do we decide on what parameters to focus on? How can we better handle and transport the fish to minimize stress?  So we get together, brainstorm ideas about how to solve a problem, arrive on a consensus, and then start working on the solution. The whole process is really exuberant and fun, even though there is no manual or guide to tell us what to do. In fact, I think it is this part of our research that makes it a valuable learning experience. In the words of Julia Giannini, who worked in the Puckett Lab last summer, “it takes some of the mystery out of the science.” Each day is different from the other but they are all tied in together by the overarching goal of our project – to examine the material and thermodynamic properties of fish schools to provide a more robust and testable benchmark for modelling collective animal behaviour.  

Our Research:

Collective behaviour are ubiquitous in social animals. Standard models of collective behaviour generally use self-propelled particles that interact with neighbouring individuals. While canonical models successfully describe qualitative features of collective structures in animal behaviour, they do not capture the dynamical behaviour of these systems in response to perturbations. The failure of models is due primarily to inaccurate interactions between individuals. However, determining these interactions from experimental data is a challenging if not impossible inverse problem as interactions are complex and stochastic.

Figure 2. All three of us at some point in our research.

In other words, collective behaviour is hard to understand, and we know very little about it. Or, in the words of our supervisor Dr. James Puckett, “we know a lot about the atom, and about stars, but then somewhere in the middle of these two scales—life, basically—it gets messy.” Our aim, broadly speaking, is to try to make sense of this mess. How does a flock of birds decide which exact turn to make? How does a school of fish arrive at a consensus and make decisions? How do people decide which way to run in case of emergencies like fire in the building where there is no way of knowing the shortest or the safest route? Most of the answers to these questions rely on how individuals interact with each other and how these interactions lead to emergent properties. In our lab, we use physics and physical laws to understand these individual level interactions. More specifically, we explore the group behaviour through the lens of physics and material science, like stretching, shearing, compressing the school of fish.

Figure2. Experimental tank with our fish school.

We observe the response of laboratory schools of negatively phototaxic (meaning they like to be in the dark spot) freshwater fish (Rummy nose tetra) to projected light fields using a high-speed camera and particle tracking set-up. Rummy nose tetras are particularly suitable for our purpose because they are extremely social, but without any social hierarchy within the group (which makes it less complicated). For our experiment, we use a light field which consist of two black boxes. Being negatively phototaxic, fish are naturally attracted towards these black boxes. Once they are in the box, we try to stretch the school by splitting the box from the center in two opposite directions.  We explored different velocities and distance of separation to see the effect on stretch parameter of the school.

Here is our sample data. The two boxes are where the dark spots are, which can be seen by the fishes but not by our infra-red camera.

Video1. Stretching a fish school by splitting the dark spot.

By now, we have collected enough data to start the analysis and calculate the school’s material properties. The analysis mainly consist of establishing that we were indeed able to stretch the group. Then we will analyse the acceleration of  the individual fish during the stretching and see if it behaves anything like what we see in physical materials like a rubber band.

Figure 4. Dr. Puckett and Aawaz.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s