Updated: Apr 26, 2021
In the past, it was thought that advanced cognitive functions, like learning, individual recognition, complex sociability and forming memories were reserved for 'higher vertebrates', such as chimps and humans. However, today, we are much more aware that many species of animals are capable of advanced cognition. There is increasing evidence that fish, such as sharks, are equally as cognitively advanced and behaviourally complex as other vertebrates.
Scientific research has shown us that sharks and rays can learn to solve cognitive tasks as well as many other vertebrates. These studies involve training sharks through classical conditioning; meaning the shark must perform a task, like pushing a button, in order to gain a food reward. When the button is pushed a stimulus will be provided, like ringing a bell. Eventually, the sharks will learn to associate the ringing of a bell with food and respond accordingly. This may sound familiar if you have heard of the 'Pavlov's dogs' classical continuing experiment, whereby dogs salivated at the sound of a bell.
For example, classical conditioning has been used to teach lemon sharks (Negaprion brevirostris) to associate a target with a food reward. Every day, at feeding time, researchers lowered a target into the water and eventually the sharks learned that that when they pressed the target with their snout they would receive food, even when no food-related stimulus (sight or smell of food) was present (Clark,1959).
Whenever something is learned, the new skill will be saved as short-term memory, but, if that knowledge is to be useful in future life, it must be stored as a long-term memory, which can be called upon in subsequent situations. The aforementioned lemon sharks retained their target-pressing skill for as much as 10 weeks after training ended. But we know very little about how long memories are retained by sharks...
In a recent study, grey bamboo sharks (Chiloscyllium griseum) were taught how to navigate through a maze to test their long-term memories. A shark was released into an arm of the maze (seen as SC1 and SC2 on the diagram) and was given a food reward when turning a specific direction at the centre crossroads. One group of sharks was trained to turn right and another group was trained to turn left (shown as F on the diagram). However, if the shark was trained by entering the maze at SC1, continuing to turn left when released from SC2 would not lead to a food reward. The maze had "landmarks" on the walls; colourful geometric shapes which could be used as reference point for location within the maze. If a reward was always gained when entering the arm of the maze with a large, blue cross at the end the shark could learn to use the landmarks to determine whether turning right or left at the crossroads at the centre would lead to a food reward (Schluessel & Bleckmann, 2012).
The bamboo sharks were able to use the landmarks to navigate within the maze. This shows us that they possess excellent spatial learning capabilities (Schluessel & Bleckmann, 2012). But what would happen after training had finished and the sharks were given a break from this task? Would they remember how to do it?
To answer this question the researchers gave the sharks breaks of several days without "reinforcement" (no exposure to the maze and landmarks or rewards in response to tasks), to see how long they could remember how to complete the task. They found that grey bamboo sharks were able to retain spatial information that they had learned for up to six weeks! They also found that, if the sharks needed a little reminder about how to complete the maze after this break, that there was no relationship between the length of the break and how much reinforcement was needed to remember. Also, the sharks completed the task just as rapidly after the break as they did when it was recently learned. This suggests that the sharks were genuinely using their memory to recall how to perform the task (Schluessel & Bleckmann, 2012).
We refer to the ability to navigate by making turns in the correct direction when using landmarks as "place learning". Place learning would be extremely beneficial to grey bamboo sharks in in their natural habitat. These sharks live in structurally complex environments, like reefs and seagrass meadows, where they feed on small benthic prey, such as worms and shellfish. Place learning would make finding food sources in their complicated habitat more efficient and would also aid in finding shelter to avoid predators... If you know there is a comfy cave to hide in behind a certain patch of coral, then you wouldn't have to waste time and energy finding a new hidey-hole when a predator appears (Schluessel & Bleckmann, 2012).
Having this advanced cognitive ability increases the sharks' "fitness" (the likelihood that an individual will go on to reproduce), because it allows grey bamboo sharks to remain safe from predators in the short-term and have increased foraging ability in the long-term (Schluessel & Bleckmann, 2012).