When we meet a group of strangers, one of the first things we’ll do is to introduce ourselves by name. Nicola Quick and Vincent Janik from the University of St Andrews have found that groups of bottlenose dolphins do something similar. When they meet one another in the wild, they exchange “signature whistles”. These whistles are unique to each individual, and they’re strikingly similar to human names. And it seems that they’re a standard part of a dolphin’s meet-and-greet etiquette.
Signature whistles were first discovered by Melba and David Caldwell in the 1960s, but we still know relatively little about how they’re used. We know that bottlenose dolphins develop their signatures when they’re a few months old, possibly modelling them on those of their mothers. They can go unchanged for decades, although males will sometimes change their whistles to resemble that of a new ally.
The signature whistles seem to act like badges of identity. One dolphin can learn information about another by listening to its whistle. But they’re not entirely like human names. For a start, they’re invented, rather than bestowed. They also convey more than just identity – they reveal the caller’s motivation or mood. “It’s a bit like in human language, where you can hear if a person sounds happy or sad, not in the choice of words they make, but in subtle acoustic features in their speech,” explains Janik.
In captivity, a dolphin will emit its signature whistle if separated from the rest of its group. This suggests that the whistles could act as contact calls, which they continuously make to stay in touch. Dolphins can even mimic each other’s signatures, perhaps just as we call each other by name. But in the wild, no one knows exactly how the whistles are used. They’re clearly common part of dolphin life – half of all the calls recorded from wild dolphins are signature whistles – but what is their role?
To investigate the role of signature whistles, scientists needed to work with free-swimming dolphins. That’s easier said than done. In the past, people have captured wild dolphins, recorded their calls, identified the signature whistles, and released the animals. How do you parse out a specific signal from the underwater cacophony of a fast-moving dolphin pod?
Janik found a way in Florida, while working with Randy Wells. Together, they noticed that the signature whistles have a distinctive rhythm. “The whistles occurring within 1 to 10 seconds of each other,” he says, “and you can use this pattern to identify signature whistles in free-swimming animals. Using the methods he perfected in Florida, Janik returned to Scotland, to work with a group of dolphins he had been studying since 1994.
Quick and Janik recorded the calls of swimming dolphin pods using underwater microphones. From 11 such recordings, they worked out that dolphin groups use their signature whistles in greeting rituals, when two groups meet and join. Only 10 per cent of such unions happen without any signature whistles. And the dolphins use their signatures nine times more often during these interactions than during normal social contact.
The signature whistles clearly aren’t contact calls, because dolphins hardly ever use them within their own groups. Mothers and calves, for example, didn’t exchange signature whistles when travelling together. And they’re not confrontational claims over territory, because bottlenose dolphins don’t have territories.
Instead, Janik thinks that dolphins use the whistles to identify themselves, and to negotiate a new encounter. The human equivalent would be saying, “My name is Ed. I come in peace.” Janik explains, “You often see ritualised behaviour sequences when animals meet. We say hello or shake hands. That doesn’t have much meaning – it’s just what you do when you meet someone else.”
Quick and Janik also found that the dolphins don’t mimic each other’s signatures when they meet up. Justin Gregg from the Dolphin Communication Project says, “In other words, dolphins are not shouting out “Hey there Jerry” to each other, they are saying “it’s me, Tim!” He adds, “We really have no idea when or why they use these whistles. This study has uncovered a brand new function for the signature whistle, which makes it rather exciting. They appear to be identifying themselves to social partners after a prolonged separation.”
But the recordings threw up a big surprise: not every dolphin exchanges its signature whistle when two groups meet up. Usually, just one member of each group does so. Quick and Janik have four possible explanations for this odd pattern.
First, it’s possible that the spokes-dolphin is the leader of the group. However, these animals don’t live in a particularly hierarchical society, and there’s no good evidence for the existence of dolphin leaders. Second, dolphins might not be very choosy about who they associate with, so they don’t need to know who they’re hanging out with. Again, this seems unlikely, since we know that bottlenose dolphins do have preferred companions. Third, the dolphins might already know who’s part of which group, so they only need to hear an individual signature to remember all the others.
Janik favours the fourth option: “it’s very specific animals that are interested in joining up”. As he says, “Just like in a group of people, not everyone wants to join up with others.”
“It would be of great interest to know whether these groups join and split from each other numerous times per day or even per hour, or if they encounter one another only rarely,” says Laela Sayigh, who studies dolphin communication at the Woods Hole Oceanographic Institution. “[That] would help to differentiate among the possible interpretations.”
That’s what Janik plans to do next. He also wants to try some playback experiments – the cornerstone of animal communication research – to see if he can provoke a specific response by playing a chosen signature whistle. He also wants to see if the dolphins ever use their signature whistles to call to others who they haven’t seen in a while.
Reference: Quick & Janik. 2011. Bottlenose dolphins exchange signature whistles when meeting at sea. Proc Roy Soc B. http://dx.doi.org/doi:10.1098/rspb.2011.2537
Photos by NASA and BBMI explorer
Bonus material: Because there’s a lot of good stuff here that I’ve trimmed for the piece, I’m including the full interview that I did with Janik, and the full statement that Justin Greggs sent to me. Lots of interesting material there for the particularly enthusiastic reader.
Interview with Vincent Janik
In what ways are the signature whistles like or not like names?
We’re also interested in this parallel partly because in humans, we can use names in a referential way. Clearly dolphins use these whistles to identify themselves but they’ll also copy each other’s whistles. We want to see if they’re using the whistles to call someone else, and we’ve got another project looking at that.
So far, all the interactions we’ve found are quite immediate. Animals copy whistles when the whistle owner starts whistling first. We don’t really have good evidence for the animals swimming out and using the calls to find somebody else. But there’s a study from Stephanie Wioodward a few years back showing that a significant proportion of dolphin whistles are the signature whistles of animals who aren’t present. That’s a tantalising bit of information. What does that mean? Are the animals looking for conspecifics and trying to call them? We don’t know yet.
One other difference is that in human names, the name has a meaning outside of identifying the person. There’s a cultural baggage too. With the dolphins making up their own whistles, that’s not the case.
Do the signature whistles carry information besides the identity of the whistler?
A study we did in the 90s showed that the whistles carry information about the motivation or the mood of the animal. It’s a bit like in human language where you can hear if a person sounds happy or sad, not in the choice of words they make, but in subtle acoustic features in their speech. You see the same thing in dolphin whistles – the overall frequency modulation pattern sounds the same, but subtle features change according to the context and carry additional information.
How long have you been studying this particular group of dolphins for?
We’ve been studying these animals as early as 1994, when I did my PhD. The dolphins we have here roam along the entire east coast of Scotland. It’s the same individuals and some of the individuals we studied then are there now. We have long histories for them and who they associate with.
Can you explain how you managed to identify the signature whistles from the recordings?
The paper that explains that is in press. Traditionally, we would somehow isolate an animal and record its most common whistle. But there’s a project in Florida led by Randy Wells, which I’m part of, where we temporarily captured wild animals, recorded them and compared those whistles to what the animals did when they were swimming freely. We found that the animals are producing these signatures in a specific pattern – it’s not just the frequency modulation, but also the rhythm of delivery. They’re occurring within 1 to 10 seconds of each other, and you can use this pattern to identify signature whistles in free-swimming animals. That’s what we did here – we applied the method that we ground-truthed in Florida.
What’s going on when two dolphins meet and exchange their signature whistles?
You often see ritualised behaviour sequences when animals meet. They go through a particular process – we say hello or shake hands. That doesn’t have much meaning – it’s just what you do when you meet someone else. So maybe it’s the same – you’re just signalling that you come in a friendly way, rather than posing a threat. It’s partly saying who you are, but also signalling that you’re coming in peace.
Why is it that only one dolphin from each group exchanges their whistles?
That’s not always the case, but there are three or four possible explanations. The most likely one I think is that it’s very specific animals that are interested in joining up. Just like in a group of people, not everyone wants to join up with others.
Alternatively, maybe there’s some kind of leader, but we don’t really have evidence for that. The social system of dolphins isn’t very hierarchical. It’s fission-fusion – they come together and split up quite regularly. So I think the chances that there‘s one leading animal are quite low.
The other possibility is that dolphins may just not be very choosy and it may not be necessary for everyone else in the group to know who else is there. I find that unlikely too because the animals do have preferred associates. I don’t think they’d form indiscriminate groups as fish do.
The fourth possibility is that we don’t know how long these animals have been separated for – we only know that it’s for on average of 19 minutes. It’s possible that they’d been together within the hour before so they have info about who was in that group. You might know who else is in the group if you hear one individual.
Do other dolphin species use signature whistles?
Yes, we have some evidence for some other species, but not in so much depth. There’s a variety of species where people have found the same stereotypes patterns – spotted dolphins, Pacific white-sided dolphins, Risso’s dolphins, humpback dolphins. But for a lot of these species, we need more evidence. By no means do all dolphins have signature whistles. Killer whales don’t, for example. If you didn’t have this fluid social structure, I don’t think you get signature whistles.
Comment from Justin Greggs
This is a great new article coming from a research organization (The Sea Mammal Research Unit at the University of St Andrews) that has been regularly churning out breakthroughs in our understanding of signature whistles. Although signature whistles were first described 50 years ago (by Melba and David Caldwell in the mid-1960s), little is known about how dolphins use them in their natural environment…
Other than the fact that dolphins produce signature whistles when separated, we really have no idea when or why they use these whistles. This study has uncovered a brand new function for the signature whistle, which makes it rather exciting. This study showed that dolphins use signature whistles when two (or more) groups greet each other after having been out of contact for a long while, which means it’s not just a simple contact call in order to continually stay in touch. They appear to be identifying themselves to social partners after a prolonged separation. That’s novel, and that’s cool.
It’s also a very strange and novel discovery that only one animal from each group seems to be producing a signature whistle during the reunion phase, as opposed to multiple animals. Why on Earth is this? It’s always a sign of great science when a new discovery brings with it new questions.
Aside from the discovery of using signature whistles before groups join after being separated, the authors describe some really great methods for obtaining and analyzing their data. One of the biggest problems for studying dolphin communication in the wild is that we rarely have any idea which animal is actually producing a recorded vocalization. Localizing is a huge problem.
Although they did not describe a method that could accurately distinguished the individual that produced the signature whistles they recorded, they were able to localize the source as coming from one group or the other with pretty good accuracy. This was more than enough info to answer their research question, and a wonderful and innovative technique. They also describe a fancy method for determining which were signature whistles and which were normal whistles from a recording where the signature whistles of the animals involved were not previously known. These two techniques made this study possible, and are pretty cutting edge.
Here are some extra thoughts on a topic I suspect might crop up in reporting on this article. There is a lot of discussion about whether signature whistles should be considered functionally referential. Because a referential signal – like a vervet monkey alarm call – could be considered something analogous to a “word” in that it is an arbitrary vocal symbol that stands in for or represents an external object or event, the signature whistle is often cited as evidence of the complex (i.e., language-like) nature of dolphin communication. But unlike alarm calls, where there seems to be a clear relationship between the call and the object/event to which it refers, it is difficult to know the extent to which a signature whistle is referential. This is partly because its function, and the contexts in which it is seen, is still so poorly understood. Although we know that dolphins have no problems at all associating arbitrary signals (vocal or visual) with both concrete and abstract objects/events, we’re still not sure if upon hearing a signature whistle, a dolphin has a mental representation of the whistler or not. We know that dolphins are more interested in the signature whistles of their close social partners, but that’s about it.
This study has contributed to this discussion by suggesting that in a social scenario where one would predict that a dolphin would need to transmit its identity to others (and thus the others would “know who is whistling”), dolphins do in fact do this. In this case, in a scenario where two groups that have not seen each other in a while are approaching each other. But a stronger case for dolphins truly using a signature whistle as a referential signal would have been if, upon two groups joining after being separated, the dolphins emitted the signature whistle of their friend whom they had not seen in a while. But this is not what happened, as the authors note: “copying does not occur when animals encounter each other initially, which is what we would predict if copied signature whistles are used to address specific individuals.” In other words, dolphins are not shouting out “Hey there Jerry” to each other, they are saying “it’s me, Tim!” This really isn’t so odd, since in this particular scenario, the animals might not have known which individuals were even in the other group, so it’s kind of useless to start addressing them if you aren’t sure who’s there. As the authors point out, the priority in this social scenario is to first identify yourself.
Because this study wasn’t really designed to ferret out the referential aspects of the signature whistle, the findings do not directly contribute to this discussion, which is likely why the authors avoid discussing this topic altogether. I suspect that some folks who read about this study are going to want to spin in in such a way that it looks like more evidence for the referential/language-like aspects of signature whistles, but that is not at all what the study was about. What it has discovered (i.e., a brand new function for the signature whistle) is certainly worth some media attention though!