The most comprehensive listening book
neuroscientist
Apple Award Winning Podcast
Podcast Episode 091: Learn how to listen with the patience of a neuroscientist

Subscribe to the podcast

Dr Alison J. Barker is a neuroscientist from Max-Delbrück Center for Molecular Medicine in Berlin.

Dr Barker teaches us about the importance of listening with patience over her multi-year study into language and dialects.

During this discussion, Dr Barker explains the importance of listening to yourself, the content and the context.

Using complex computer algorithms to listen, Dr Barker discerned patterns in language faster than a human. She explains the limitations of software and where humans hold an advantage.

Transcript

Podcast Episode 091 – Learn how to listen with the patience of a neuroscientist

Oscar Trimboli:

What does an American neuroscientist studying in Berlin, the Game of Thrones and deep listening all have in common? Well today, you’re going to find out. Deep listening impact beyond words. Good day. I’m Oscar Trimboli, and this is the Apple award-winning podcast, Deep Listening, designed to move you from a distracted listener to a deep and impactful leader. Did you know you spend 55% of your day listening, yet only 2% of people have ever been taught how? In each episode, we explore the five levels of listening. Communication is 50% speaking and 50% listening, yet as a leader, you’re taught only the importance of communication from the perspective of how to speak. It’s critical you start to build some muscles for the next phase in how to listen. The cost of not listening, it’s confusion. It’s conflict. It’s projects running over schedule. It’s lost customers. It’s great employees that leave before they want to. When you implement the strategies, the tips and tactics that you’ll hear, you’ll get four hours a week back in your schedule. I wonder what you could do with an extra four hours a week.

Oscar Trimboli:

Dr. Alison Barker, originally from Rhode Island in the United States moved to Berlin in Germany to study one of the Earth’s most unique mammals, the naked mole-rat. The naked mole-rat is native to some of the harshest conditions in the world, originally from East Africa. Naked mole-rats are mammals and they get their naked name from their lack of fur, and they’re a focus of study because they’re cold-blooded mammals, quite an unusual combination. You’re probably wondering what’s all this got to do with deep listening. The naked mole-rats have survived and evolved because of their highly evolved and cooperative colonies, and it’s their ability to listen to nuanced communication and dialect that have helped them sustain themselves over many decades and centuries. We explore how Dr. Alison Barker spent nearly two years listening to the naked mole-rats and used microphones, as well as computers and software, to listen to their language and their dialogue. Why I chose to explore this discussion? Well, it’s about listening at level two. It’s about listening to the content, and it’s also at level three, listening for the context. Let’s listen to Alison.

Alison Barker:

My name is Alison Barker. I’m a neuroscientist at the Max Delbrück Centre for Molecular Medicine in Berlin, and I’ve been studying how vocal communication works in naked mole-rats.

Oscar Trimboli:

What do you struggle with when it comes to your listening?

Alison Barker:

I can be quick to think I already know the answer. Especially in the scientific context and through academia, you’re trained to interject quickly and kind of share your thought process. I have to say living abroad has really helped me with that. I find whenever I speak German or I’m in a group with Germans, I’m much more of a listener than a talker.

Oscar Trimboli:

How long did you study naked mole-rats? Because one of the things you had to learn was how to be patient in your listening. How did listening to naked mole-rats change the way you listen?

Alison Barker:

Studying animal communication is fundamentally a listening exercise, and you have to maybe reconsider what it means to listen, and it requires a different type of patience. I spent a lot of time just standing there with a microphone. When I first met the naked mole-rats, and this was about four years ago, the first thing I noticed about them was not their characteristic features, which I think a lot of people harp upon. They’re a bit unique in the way that they look, but actually what I noticed is that they were constantly chirping. So they’re constantly vocalising. And I was really fascinated by that question, what are they communicating? Because I had no idea what it meant, and they actually sounded like birds. A lot of patience and a lot of necessary observation to try and not understand just what they were actually saying, but the context for their whole societies. When they communicate, they’re interacting in a different way than maybe we are, and I had to leave some of my human bias behind. You have to understand sort of a new culture with these animals.

Oscar Trimboli:

Zoom us into the lab listening to the naked mole-rat individually. Or how the colony’s set up, so you listen to them in groups?

Alison Barker:

We recorded from them individually. We’d take them out and put them in a little chamber with a microphone, and that was really essential for us to get enough information about the individual voice. We could use that to train our machines. I also did a lot of listening to the whole colony. They’re highly cooperative creatures. They live with sort of unique social structure for mammals. Much like bees or ants, they have a single queen, and in our lab, the largest colony was about 50 individuals. We housed them in these interconnected chambers, like you might see in an ant farm, and getting a microphone in there was a little bit complicated because we keep them in a very hot and humid conditions. I really had to train a bit too to be in there for long periods of time. We would set up several microphones within different chambers and we could kind of watch them and listen to them at the same time.

Oscar Trimboli:

Take us back to the origins of this research. What started the research and how did it evolve?

Alison Barker:

The naked mole-rats were first discovered in the early 1800s in the field, and in the 1980s, they were brought into the lab. And a lot of people noticed these really cool biological features that they’re cold-blooded. They live for a long time. They live in these elaborate social groups. And I think most people that encountered these animals often would say they’re making some weird noises and we’re not sure what that means. This was kind of in the literature, but no one had really dove into it. About four years ago, I was in Berlin and I met with Professor Lewin, who is at Max Delbrück Centre, and he had had the naked mole-rats for about 12 years and he was studying how they respond to different types of painful stimuli and how they respond to hypoxia. Then I chatted with him and then I met the mole-rats, and I kind of just thought, “Okay. This is it. This is what I have to do.” For me, the question was really, what are these vocalisations doing? Are they contributing to their organisation as a social group? Because we know they’re super cooperative, but no one had really figured out how that cooperation is organised. We have about 200 and they’re spread across nine different colonies.

Oscar Trimboli:

And how do you name the colonies?

Alison Barker:

When I joined the lab, they just had an identifying four-digit number. We decided that if I was going to be spending a lot of time with them, I should get to know them a bit more. We were both sort of fans of the Game of Thrones series, which actually is a bit of a violent show. When you understand the naked mole-rat, lifestyle a bit… I mean, it is very dependent on the queen and she has a lot of power and a lot of control over the colony. We decided we would name them after Game of Thrones. I took a week or two where I just spent the whole day, the whole weekend with the naked mole-rats, and in the evenings, I rewatched the first season of Game of Thrones. And then I made an emotional decision about how to name them. We have Targaryen, Stark. Game of Thrones is kind of loosely inspired by the English tradition of having different houses and different royal families. I felt a bit bad in the end because some of the events that transpired in the colonies I felt like were very actually similar to what happened in the Game of Thrones. There were some overthrows. There was a battle at one point.

Oscar Trimboli:

What were the various characteristics of each colony?

Alison Barker:

House Targaryen actually started very small. It was newly founded, and they seem especially aggressive. In the end, they actually expanded, and now they’re one of our largest colonies. I think that fit quite well with the beginning arc of the show, where House Targaryen expanded its rule over part of Westeros. We had one colony that only had two individuals and I called that House Greyjoy because it was like an island nation. But yeah, this is very whimsical, I think.

Oscar Trimboli:

Thinking about you spending time with microphones and patient listening over hours and days and weeks and months and years. What did you notice you were listening for and how did that change over time?

Alison Barker:

I think at first I wasn’t even really aware of what I was listening to because everything was different and it was such a multisensory experience. I’m also watching the animals and listening and also having to adjust to the environment of standing there with the microphone and the heat and how to hold my hand. And so there was just a lot of things happening at once, but slowly, I was able to sort of filter some of those other things out, and I was able to start really listening, really noticing individuals having sort of different voices and how individuals would interact with one another and how the sounds were sometimes different. I also was guided a bit in this by… There was one paper that was published in the early ’90s, which was sort of the first naked mole-rat dictionary. It had a little bit of knowledge of what I was looking for, and in the end, we identified some new sounds as well. Now, I know a lot of the animals very well and I can recognise them by voice.

Oscar Trimboli:

I think over time, we, as listeners, do pattern recognition, and eventually you had to code these patterns mathematically into a computer model and allow the computer model to start to listen as well.

Alison Barker:

What we’re doing is we’re giving the machine some information and then we’re asking it to recognise patterns and then use those patterns to predict future outcomes. What I like about that is that’s how the brain works as well. It’s also a bit of a black box. So we’re given a lot of sensory information. It goes into our brain or into a machine. And then there’s some sort of pattern recognition that happens, and that allows us to make predictions for future sounds or recognising things that we haven’t been exposed to before and classifying them correctly. We use the machine learning in two ways. The first thing we did, which was really kind of a game changer, was that we used it to automate the classification of sounds so we could record very long recordings and we could segment the information of individual sounds and we could classify which type of sound it was.

Alison Barker:

This allowed us to get a lot of data, and that’s really important when you want to train a machine or train a human. You need to have a lot of good quality data. And so in the past, this was a limiting factor. So we were able to train some algorithms to really rapidly parse out the sounds. Then we decided we would take eight different features. The eight features are the asymmetry of the soft chirp, the duration, the height, the peak frequency, the slope, the pitch, a feature called Wiener entropy, which describes the messiness of the sound, and then finally, zero-crossings, which is another measurement of the frequency of the sound. And just give one type of machine these eight features, and we purposely chose a very simple sound, which looks basically just has an upsweep and then a downsweep, which is pretty symmetrical.

Alison Barker:

We chose a very simple sound that we knew the naked mole-rats used to greet one another, and we took these just eight very simple features and we asked if the machine could then use that to make predictions about where an individual came from or what colony it came from. And I think the basic idea of this was that if there’s some information in those sound features that may be correlated to say the identity of a colony, then the machine would be able to learn that, but if there was no underlying organisation to the data, then the machine wouldn’t be able to learn that, and we kind of use that as a tool to help us guide our future research direction.

Oscar Trimboli:

Everybody thinks machines do things faster and better than humans. What did you learn about the ability for the machine to listen compared to you?

Alison Barker:

It is true in one sense that machines can certainly do some things faster than us. The machine could integrate much more data than I could very quickly, but the machine was limited in what it could classify because I controlled what I put into the machine. It could only learn the choices that I gave it. We gave eight features, and the machine could only use those features and they could only predict one of four or one of five colonies, because that was the only data that it had. There are some sort of qualitative things that I was able to learn, and I’m still not sure how I learned them, that I could do better than the machine, and I think that this is… I mean, it is quite amazing. Machines are getting very good at voice recognition, image recognition, but they still don’t perform as well as humans, and we don’t really know why that is. What it tells us on a very broad level is that the human brain is a really remarkable machine and it has been shaped by things that we can’t even imagine through our evolutionary history and through our experience. This is one of the reasons I love being a neuroscientist is we get to play with this and try and take it apart.

Oscar Trimboli:

Colonies have dialects, and I’m assuming the dialects are driven by the queen.

Alison Barker:

Mm-hmm (affirmative). How does a dialect emerge even in a human society, there’s obviously a shared environment and there’s a shared sound environment, and then there’s some level of cultural reinforcement where you obviously want to belong to the social group, and so you kind of want to adhere to certain mores within the group. And for the naked mole-rats, the vocal culture is something that’s important for them to indicate belonging and to exclude individuals that don’t belong to the colony, and this is very important for them in terms of resource allocation. One hypothesis would be that there is actually one sort of template that everyone in the colony is matching to, and this could be coming from the queen, but we actually didn’t really have a way to test this because if you remove the queen from the colony, the whole colony falls apart. And so we really didn’t want to do this, but it sort of happened accidentally, and this is something that’ll be interesting for the Game of Thrones fans is that we actually had a queen that was murdered in House Stark. And actually, this happened twice where the queen was lost.

Alison Barker:

And so I had been, again, sort of patiently tracking these animals over multiple years. And I was at first quite devastated when the queen was murdered, because I thought we would lose all this data, but in the end it turned out to be a great thing because we had a period of time where we had a stable queen and we had recordings of the dialect, and then we had subsequent periods of what we called anarchy, where the social organisation fell apart. And what we were able to see is that the dialect actually also disintegrated in these periods without a queen. We thought that perhaps everyone was matching their template dialect to the queen, and actually this turned out not to be precisely the case. We still don’t understand how the queen controls the dialect. We know that not everyone is perfectly matching to her, but she’s somehow behaviourally or perhaps hormonally enforcing this convergence of a common dialect within the colonies.

Oscar Trimboli:

How different are the dialects?

Alison Barker:

This is always a fun party game. I can find them to be quite distinct. I talked about these eight features before. We found that two of them were actually sufficient for the animals to recognise the differences between colonies. So we created some artificial sounds, just looking at two features. The frequency, for the listener, you could think of how high a sound sounds or how low, and this is kind of similar to pitch. For example, men have a lower pitched voice than women, and you can hear this in the frequency of the colony sound. We could also see the big difference between asymmetries. That’s the frequency that the sound starts at and then that the sound ends at. There’s sort of a difference in that. If you train yourself a bit, you can recognise. It’s not incredibly obvious at first listening.

Alison Barker:

For me, the best way that I describe this is when I tried to learn German, I could not hear the difference between the umlaut and the non-umlaut sounds. And really for several years, it just always sounded the same to me, and now by some small miracle, I can really hear the difference. And so I think for the average human perhaps, listening to the naked mole-rat dialects is kind of similar. The difference is there, and once you know it, it’s really clear, but it’s maybe not so obvious at first listen.

Oscar Trimboli:

Does dialect evolve over the generations of naked mole-rat colonies?

Alison Barker:

Naked mole-rats are quite long-lived. In fact, the oldest member in our colonies is 25 years old, and for a rodent, that’s pretty remarkable. This question of setting them over generations, they may outlive me and my research to really study this. I can say though, that there does seem to be some drift in the dialect. We noticed this in House Stark when we had periods with different queens. While the dialect became much more cohesive and much easier to recognise, it was actually not the same as the original dialect. So there was some drift. It seems that maybe sort of the dialect of the colonies is an active experience. So animals can’t just sort of memorise it. They have to be constantly engaged in practising it and listening and sort of modifying their own dialect to sort of match the members of the colony. So I think it’s much more of an active process.

Oscar Trimboli:

Tell us more about the survival of each colony and why xenophobia is so important.

Alison Barker:

In what we know from the field work of naked mole-rats, they are in fact very wary of foreigners because they have very limited resources, and I think one of the reasons they’ve sort of survived in these harsh, arid environments is by banding together and being highly cooperative and sharing food. And so we know that individuals, a subset of sort of worker individuals will go out and forage food and actually bring it back and share it with the colony. It makes a lot of sense, and it’s actually very important to know who is part of your group and who is not so that you can share these limited resources. This is one aspect where we think the dialects can help with this. They can sort of help exclude foreigners. The flip side of this is that the dialect helps them be very cooperative, and we only looked at one sound, and the naked mole-rats actually make more than 20 different sounds. What one should also take away from this is that vocal communication is really important for cooperation. As we delve into these other sounds more, we’re going to realise that that’s what helps make them so cooperative and so successful in a really difficult natural environment.

Oscar Trimboli:

Alison, when it comes to listening, what’s the naked mole-rat taught you about listening?

Alison Barker:

It’s taught me to be very patient and to not have a very strict idea of what communication is. Vocal communication can take on many forms and it can be subtle. It can have other behavioural cues that are associated with it. I really needed to abandon my own biases of what communication was, and that was coming from a human perspective to really kind of understand the naked mole-rats. What we can learn from them is that when you’re in a difficult situation, being able to communicate with members of your social group is really important to survive in that situation.

Oscar Trimboli:

What happens when you join a new team, a new project or become a member of a new group? What do you notice about the way people speak? And does that group use jargon or dialect in the way they communicate to each other? How do you notice language patterns in the group? For me, it’s often about, where’s the gravity of the conversation? Are they speaking consistently about the past, the present or the future? Or do they speak very internally focused, or are they looking outside? Do they speak in stories or do they speak in details? All of these are clues about patterns when it comes to listening. Another way you could think about it is, do you listen for similarities, or do you listen for differences when you move into a foreign environment? You might be joining a new project. You might be joining a new team.

Oscar Trimboli:

Whether you listen for differences or similarities, neither is right or wrong. The question is which one’s appropriate for the situation you find yourself in right now. For most of us, our default listening orientation is listening for similar patterns. We do this because it’s the fastest way. It’s a shortcut for our mind to make sense of our current context, especially when the environment is one that we’re not really used to. Alison used a really powerful technique, a technique that’s used well at level three, when we’re listening for the context. To make sense of the mountains of data, the rats, the colonies, the cultures, the hierarchies, the queens, and everything around her, Alison used a really simple technique. It’s a mental shortcut to help, to simplify and to unify when it comes to dealing with foreign and complex issues. Alison used the metaphor that is the Game of Thrones. And when you can explain a relatively new concept using the power of a metaphor, the idea seems to make sense so much faster for those who are listening to you. Equally, it makes it easy for you to make sense of the speaker.

Oscar Trimboli:

If you simply ask them the question, if this was a movie, what would it be? If this was a book, what would be on the cover? If this was the subject line in an email, what would it say? If this was a sticker on the back of a car, what would it say? All of these help the other person to simplify and clarify their idea. Some of my other fun ones are, if this was an animal, what animal would it be? Or I’m curious, what colour does that feel like for you right now? And no matter what colour they say, the next question is always, well, what colour should it be? Sometimes, I say the same colour. Sometimes, it’s different. It really doesn’t matter. This technique is used to clarify really quickly for everybody, a shortcut to make sense of complexity.

Oscar Trimboli:

What I took away from today is that listening requires patience. Whether it was Alison learning to speak German or whether she was learning to speak naked mole-rat, I think what she showed is that by deliberately pausing, being patient and taking the time to listen not only to the obvious, but also to the nuance in there as well. I’m Oscar Trimboli, and I’m on a quest to create a hundred million deep listeners in the world, and you’ve given me the greatest gift of all – you’ve listened to me. Thanks for listening.

 

Subscribe to the podcast