Post provided by Catarina Vila Pouca

This post is also available in Portuguese

Hello there! My name is Catarina Vila Pouca and I study how and why animals behave and learn in different ways. I have had a passion for sharks and swimming for as long as I can remember, and so in my career I have mostly focused on sharks and fish. My latest project, together with colleagues and students from Wageningen University and The University of West Indies, is described in a recent Methods in Ecology and Evolution paper. It started with a simple, but unanswered question: How can we study fish learning abilities outside our usual laboratory settings?

Challenges in studying fish cognition in the wild

It has been well documented that captivity can impact behaviour and cognition (learning, memory, decision-making) and bias our understanding of how animals learn and make decisions in their natural environment. Yet, the vast majority of fish cognition research to date (including my work) comes from laboratory studies with captive animals. This is because it is difficult to design, and more importantly, to carry out experiments with fish in the wild. Discussing this back in 2019 with my advisor at the time, Alexander Kotrschal, we thought it was time to start developing an apparatus that would allow us to run learning assays similar to what we did in the lab, but in-situ in the field.

Finding inspiration and test, test, test!

The challenge of taking behavioural and cognitive tasks to the wild is not exclusive to fish, and so we started with a review of the literature on wild cognition to find inspiration for designs that could work with fish, and that could be comparable to the tasks that are typically run in the lab. Testing these designs involved a lot of trial and error, no surprise there! I started with trap-style designs, based on commonly used fish traps. These usually have a small funnel-like entrance that makes it hard for fish to get out of the trap; we thought it was a good idea to adapt these traps to lead fish into a choice chamber, and to add another funnel-like exit route if they chose the correct side. While not a complete fluke, these designs had many downsides: the fish were not thrilled to be “trapped” in the choice chamber, and when trying it in the field, they were very hesitant to approach these trap-boxes. Plus, this type of task would not be easily adapted for many cognitive domains (like numerical learning, memory, problem solving), so we decided we needed a new design.

The winning design

Our main source of inspiration for the new design was the work of Rachael Shaw on wild New Zealand robins, which uses a foraging board with pivoting discs attached to the board with a screw. Now, the use of foraging boards with displaceable discs was already a standard lab assay with many fish such as guppies, cichlids, or zebrafish – but in these lab tasks, the discs were loose. The novelty in Shaw’s apparatus was the addition of a screw to attach the discs to the board, which prevents losing them when running the task in the wild (very important!). Another advantage of such a foraging board with pivoting discs is that it can be customised and adapted for different learning and memory tasks, as well as used for social foraging experiments. By varying the colour of the discs, or by painting shapes or dots into the discs, we can examine different cognitive domains.

Similar to the New Zealand robin studies, we also decided to build our foraging board with several discs – this meant that we could collect data for several ‘runs’ of the task with just one presentation of the foraging board, therefore maximising data collection while minimising disturbance of fish populations. After some pilots in the laboratory to make sure data collection and analysis was possible with this design, we took it for a spin with wild guppies in beautiful Trinidad and Tobago. Because we wanted to examine not only learning but also social foraging dynamics, our guppy tests were done with wild-caught fish that were individually marked. In the paper, we present some cool results on group learning curves, but also on individual proficiency in dislodging the discs – and show that some guppies are a bit lazy and prefer to scrounge for the rewards of others.

The final step was to think of how we could take our foraging board to different habitat types. The board by itself can be easily placed in shallow streams or ponds, but we wanted to be able to access deeper bodies of water. Enter our placement structure: a heavy base and a sliding top structure that contains the foraging board and video camera. The remote deployment was successfully tested in a natural pond with a population of sticklebacks, which quickly interacted with the board and learnt the task.

Want to use our foraging board? Here are some tips

We hope our foraging board will be used to expand fish cognition research to other species and environments and lead to significant advances in the field! Before you dive in, here are some tips:

– There is a fine balance between making the discs too lose (they might come open when you place the board or with the current) or so tight that the fish cannot open them. Pilot this with your species.

-We applied a thin layer of petroleum jelly (e.g., Vaseline) to the reward holes with a cotton swab, to make sure the rewards would not float.

-Air bubbles inside the reward hole can make it difficult for the fish to dislodge the disc. One way to prevent this is to add water to the holes when you prepare the plate.

Feel free to reach out if you have any questions, concerns, or simply want to share your own foraging board adventures (in which case please send photos as well!).

Post edited by Sthandiwe Nomthandazo Kanyile