Longest Running Wild Dolphin Study In Florida Reveals Complex Social Communication Through Unique Whistles

Human fascination with bottlenose dolphins goes back thousands of years, at least as early as Greek mythology.

But it wasn’t until the 1960s that methodical research into dolphin communication began. Scientists like John Lilly and the husband-and-wife team of Melba and David Caldwell tried various experiments to decipher the sounds dolphins can make.

The Caldwells figured out a way to record isolated animals in human care. They discovered that each individual dolphin communicated mostly with one unique whistle, which they called the “signature whistle.” Researchers now know that these whistles convey identities much like human names do. Dolphins use them to stay in touch with each other in their murky habitat, where vision is limited. It’s like announcing “I’m over here!” when someone can’t see you.

This discovery is foundational to my own research. I’ve been studying communication in wild dolphins since the mid-1980s, when I joined my mentor Peter Tyack in documenting signature whistles in wild dolphins for the first time. Our team’s research focused on a resident community of free-ranging bottlenose dolphins in waters near Sarasota, Florida, where I continue to work today.

This collaborative study, led by Randall Wells of Brookfield Zoo Chicago’s Sarasota Dolphin Research Program, involves numerous researchers from a variety of institutions, who study different aspects of dolphin biology, health, ecology and behavior. Begun in 1970, this is the longest-running research project on a population of wild cetaceans – whales, dolphins and porpoises – in the world.

Recording and observing

Researchers know the age, sex and maternal relatedness of almost all of the approximately 170 dolphins in the Sarasota community. This depth of knowledge provides an unprecedented opportunity to study communication in a wild cetacean species.

The dolphins in the Sarasota project are periodically subject to brief catch-and-release health assessments, during which researchers, including me, briefly handle individual dolphins.

Our team attaches suction-cup hydrophones directly onto each dolphin’s melon – that is, its forehead. We then record the dolphins continuously throughout the health assessments, taking notes on who is being recorded when, and what is happening at the time.

This is how my colleagues and I were able to confirm that wild dolphins, like captive animals, produced large numbers of individually distinctive signature whistles when briefly isolated from other dolphins. Through observations and recordings of known free-swimming dolphins, we were further able to confirm that they produced these same signature whistles in undisturbed contexts.

We have organized these recordings into the Sarasota Dolphin Whistle Database, which now contains nearly 1,000 recording sessions of 324 individual dolphins. More than half of the dolphins in the database have been recorded more than once.

We identify each dolphin’s signature whistle based on its prevalence: In the catch-and-release context, about 85% of the whistles that dolphins produced are signature whistles. We can identify these visually, by viewing plots of frequency vs. time called spectrograms.

Signature whistles and ‘motherese’

The Sarasota Dolphin Whistle Database has proved to be a rich resource for understanding dolphin communication. For instance, we have discovered that some calves develop signature whistles similar to those of their mothers, but many do not, raising questions about what factors influence signature whistle development.

We have also found that once developed, signature whistles are highly stable over an animal’s lifetime, especially for females. Males often form strong pair bonds with another adult male, and in some instances, their whistles become more similar to one another over time. We are still trying to understand when and why this occurs.

Dolphin mothers modify their signature whistles when communicating with their calves by increasing the maximum frequency, or pitch. This is similar to human caregivers using a higher-pitched voice when communicating with young children – a phenomenon known as “motherese.”

Also similar to humans is how dolphins will initiate contact with another dolphin by imitating their signature whistle – what we call a signature whistle copy. This is similar to how you would use someone’s name to call out to them.

Our team is interested in finding out if dolphins also copy whistles of others who aren’t present, potentially talking about them. We have seen evidence of this in our recordings of dolphins during health assessments, which provide a rare context to document this phenomenon convincingly. But we still have more work to do to confirm that these are more than chance similarities in whistles.

Shared whistle types

Another exciting development has been our recent discovery of shared whistle types — ones that are used by multiple animals and that are not signature whistles. We call these non-signature whistles.

I could hardly believe my ears when I first discovered a repeated, shared non-signature whistle type being produced by multiple dolphins in response to sounds we play back to them through an underwater speaker. We had previously believed that these non-signature whistles were somewhat random, but now I was hearing many different dolphins making a similar whistle type.

Our team originally had been using the playbacks to try to determine whether dolphins use “voice cues” to recognize each other – similar to how you can recognize the voice of someone you know. Although we found that dolphins did not use voice cues, our discovery of shared non-signature whistle types has led to an entirely new research direction.

So far, I’ve identified at least 20 different shared non-signature whistle types, and I am continuing to build our catalog. We are hoping that artificial intelligence methods may help us categorize these whistle types in the future.

To understand how these shared non-signature whistle types function, we are carrying out more playback experiments, filming the dolphins’ responses with drones. We’ve found that one such whistle often leads to avoidance of the drones, suggesting a possible alarm-type function. We have also found that another type might be an expression of surprise, as we have seen animals produce it when they hear unexpected stimuli.

More difficult, more interesting

So far, the main takeaway from our experiments has been that dolphin communication is complex and that there are not going to be one-size-fits-all responses to any non-signature whistle type. This isn’t surprising, given that, like us, these animals have complicated social relationships that could affect how they respond to different sound types.

For instance, when you hear someone call your name, you may respond differently if you are with a group of people or alone, or if you recently had an argument with someone, or if you’re hungry and on your way to eat.

Our team has a lot more work ahead to sample as many dolphins in as many contexts as possible, such as different ages, sexes, group compositions and activities.

This makes my job more difficult – and far more interesting. I feel lucky every day I am able to spend working on the seemingly infinite number of fascinating research questions about dolphin communication that await answers.