
Animals may not have musical instruments, but the way that some species form complex patterns of vocal sequences is remarkable. A recent study has found that Antarctic leopard seal. mating call sequences seem to be similar in predictability to human nursery rhymes.
In some cases, the vocal sequences an animal makes has all the elements of song. A song is made up of repeating sounds that follow a set pattern. These sounds can be combined into notes or syllables, which are then grouped into phrases or motifs.
The most prominent example of animal song is found in songbirds, which include some of our common garden birds, like the robin, the blackbird, or the wren. However, songs can also be found in other animals, ranging from small animals like fruit flies, to mice and various bats, to gibbons and even to humpback whales.
Human music and song appear to primarily facilitate group cohesion and coordination. They can also function in our courtship, as exemplified by the serenade, which in its original form was a night-time song used to honour and court a potential partner.
In animals, song serves multiple purposes and is often tied to mate attraction (male fruit flies “sing” with wing vibrations to court females) and mate stimulation (male ultrasonic song in mice has been linked to ovary development in females). Is is also linked to competition, because male songbirds also use song to proclaim and defend their territories against potential rival males.
Although some of the characteristics of animal songs have a lot in common with human songs, they are not made of series of words that each have a specific meaning. Research on vocal sequences and song in animals is therefore mostly focused on identifying patterns in animal vocal communication and studying what function a song or vocal sequence might serve.
Within this context, the new study on Antarctic leopard seals made some interesting discoveries. The study, which was published in the journal Scientific Reports, analysed underwater recordings of 26 male leopard seals from eastern Antarctica.
The researchers used a single hydrophone (a microphone that allows sound recordings in aquatic environments) in the Davis Sea, Antarctica, to record leopard seal vocalisations over the span of six years. Leopard seals are the second-most important predator in Antarctica, almost at the top of the food chain. The orca is the only predator that is known to hunt these seals.
Antarctic leopard seals are usually solitary, except during the breeding season, and each need extensive hunting grounds. However, this means that communication between seals needs to carry over long distances.
Male leopard seals emit loud underwater sounds during the breeding season from November to January. This type of call is called a stereotyped sound, because they can be put into distinct categories.
Scientists believe these calls serve multiple purposes. First, they are thought to be used to attract the attention of females and second, they are thought to be a deterrent for other males. Females seem to also make broadcast calls but only during a short time window when they are sexually receptive.
The males recorded in this study used five different stereotyped calls and strung them together into elaborate series. Although they shared the same repertoire of stereotyped calls, the order in which they strung them together into sequences varied between seals.
The researchers also explored whether these leopard seal vocal sequences follow a predictable sequence pattern and if so, how this compared to sequence structure in other mammals (including humans). Even though the authors labelled the sequences “songs”, they did not mention whether or not they meet all the criteria needed to qualify for song (as described above). I will therefore refer to them here as vocal sequences.
Using several different statistical approaches that aim to assess entropy (randomness) in these sequences, they found that these seal vocal sequences appear to follow predictable patterns. Nonetheless, predictability for leopard seal sequences was lower than for songs of humpback whales and whistle sequences of bottlenose dolphins.
When compared to humans, the authors found that leopard seal sequences were less random than music by classical and romantic composers or the Beatles, and that they were similar to the randomness found in nursery rhymes.
The authors further found that particular sequences were used by the same males over several consecutive days and speculate that these sequences might be used as a kind of acoustic signature. This has been found for series of clicks, called codas, in sperm whales, and has been suggested for the songs of some songbirds, like the chiffchaff.
However, in most other species, identity seems to more commonly be encoded in single vocalisations, such as the signature whistles of bottlenose dolphins or the vocal signature calls of parrots.
The study on leopard seals provides some fascinating insights into commonalities and differences between human music and the sequential vocalisation patterns of certain animals.
But it also demonstrates the difficulties scientists face in trying to identify the rules that govern animal vocal sequences and to decipher the messages encoded within them. For example, we usually require a huge number of recordings from a species. Obtaining these recordings can be relatively easy for species that are common where lots of people live. But it can be difficult for animals that live in remote areas.
This study used mathematical equations to analyse the recordings that seals made. However, many animal researchers are beginning to use AI to help find patterns in large databases. I can only imagine what new and exciting discoveries in research on animal communication patterns might be just over the horizon.

Rüdiger Riesch does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.
This article was originally published on The Conversation. Read the original article.