What is it about?
We tested cleaner wrasse in a classical reasoning paradigm called transitive inference (i.e., if A is bigger than B and B is bigger than C, then A is bigger than C). We used 5 plates that differed in color and pattern, called A, B, C, D and E. The fish learned to prefer A over B, B over C, C over D, and D over E, using a food item as a reward when the fish chose the correct plate. Once they learned those, we tested whether they would prefer: A over E (as a control, those are the anchors and the fish should not have any problem as A was always rewarded in the learning process and E was never) and B over D. The later is the transitive pair as B and D plates had been both rewarded and not rewarded in the learning process. Fish succeeded AE pair without any problem but BD was not solved above chance. As this could not be explained by any potential experimental bias, we compared our methodology with a study that was previously published. The only relevant difference that came out was the difference in training methodology. While our fish learned the 4 basic pairs (AB, BC, CD, DE), theirs also learned non-adjacent pairs et sessions were repeated and controlled multiple times. We consider that this might be the key to their success, but we also claim that this training methodology is taking away the ecological relevancy. Indeed, transitive inference appears to be a tool to deduce unknown pair relationships from a minimal amount of informations. Thus, the need to train the fish over and over again seem to demonstrate that cleaner do not possess the ability naturally but that learning can overpass that, at least in this foraging context in the lab.
Featured Image
Why is it important?
It is important to investigate another potential component of cleaner wrasse cognitive toolbox. We think that transitive inference is not an important ability for a cleaner wrasse in nature. Therefore, we wanted to test whether they would possess the ability anyway as it had been found in species with both bigger and smaller brains (insects and mammals and birds).
Read the Original
This page is a summary of: In the absence of extensive initial training, cleaner wrasse Labroides dimidiatus fail a transitive inference task, PLOS One, June 2023, PLOS,
DOI: 10.1371/journal.pone.0287402.
You can read the full text:
Contributors
The following have contributed to this page