From Our Neurons to Yours

Why voices light us up—but leave the autistic brain in the dark | Dan Abrams

• Wu Tsai Neurosciences Institute at Stanford University, Nicholas Weiler, Dan Abrams • Season 7 • Episode 11

Share episode

Recognizing a familiar voice is one of the brain’s earliest social feats. But what are the brain circuits that let a newborn pick out mom in a crowded nursery? How do they change as kids turn toward friends and the wider world? And what are we learning about why this instinct fails to develop in the autistic brain?

This week, host Nicholas Weiler joins Stanford neuroscientist Dan Abrams on the quest to understand the neural “hub” that links our brains' hearing centers to the networks that tag voices as rewarding, social, and worth our attention. The findings could reshape early-intervention strategies for kids on the spectrum.

Learn More

We want to hear from your neurons! Email us at at neuronspodcast@stanford.edu

Send us a text!

Thanks for listening! If you're enjoying our show, please take a moment to give us a review on your podcast app of choice and share this episode with your friends. That's how we grow as a show and bring the stories of the frontiers of neuroscience to a wider audience.

Learn more about the Wu Tsai Neurosciences Institute at Stanford and follow us on Twitter, Facebook, and LinkedIn.

Nicholas Weiler (00:11):

Welcome back to From Our Neurons to Yours from the Wu Tsai Neurosciences Institute at Stanford University, bringing you as always to the frontiers of brain science. A few years ago, Dan Abrams and his team at Stanford published a study that made headlines, though honestly, a lot of those headlines miss the point.

(00:33):

The story went something like this, neuroscience says teenagers don't listen to their mothers. Great. Well, what Dan's research actually showed was something a little deeper than that and more revealing. At the heart of the study, it was a question about how children respond to different voices as they develop.

(00:54):

When young kids hear their mother's voice, their brains don't just process the sound, they light up across brain circuits tied to reward and social meaning, and that makes a lot of sense. This goes back to early, early development. There are studies that show babies recognize their mother's voice even in utero.

(01:13):

Now, what Dan and his team found is that in adolescence, that response shifts, the teenage brain starts tuning in much more to unfamiliar voices. So it's not that teenagers stop listening to their moms, it's that the voice is a map of our social connections. And during adolescence, that map is redrawn to focus on a broader social universe.

(01:37):

This matters because the voice isn't just another sound. Human voices form the fundamental building blocks of social connection. And when the brain's systems for processing voices don't develop as expected, the effects can cascade, shaping language, learning, the ability to connect with other human beings.

(01:56):

This is the question we're going to tackle with Dr. Abrams today on the show because what Dan and his team are really trying to get at is whether early problems with the fundamental development of how our brains process voices might contribute in surprising ways to autism and other social developmental challenges.

Dan Abrams (02:15):

So one of the first signs that a child may have autism is associated with voice processing, and this is widely recognized in the field. Lots of parents first sense that there may be something amiss when their child does not respond to voices, including when someone calls their name.

Nicholas Weiler (02:34):

And that can be from a very early age, like two?

Dan Abrams (02:37):

From even younger than that, even in their first year of life, these kinds of voice processing challenges can sometimes be evident, and neuro-typical kids are really tuned into voices. Anyone who's ever spent time with an infant, of course, they have no idea what you're saying, but they'll watch two people having a conversation, and their head will kind of like watching a tennis match or something.

(03:03):

They'll shift from one speaker to the next. They are interested in voices, and we think that's this kind of initial attraction to voice is an important aspect of social development. This attraction helps them to begin to discriminate different vocal sounds, speech sounds, which they eventually will learn to mimic through babbling and then ultimately start to attribute meaning to those sounds.

(03:27):

We think this initial attraction to vocal sounds is this key aspect of social development, and the voice is just primitive signal, like where an infant hears their mom's voice. We know that a fetus is in utero can identify their mom's voice. So the neurotypical brain is highly attuned to specific voices from a very early stage of development.

(03:54):

And then in kids with autism, we often see something quite different where kids with autism will tune out from the voices in their environment. Parents often think there may be a hearing impairment. They often wind up at an audiologist's office to have their peripheral hearing checked to see if there may be some peripheral hearing impairment that could preclude them from tuning into voices.

(04:16):

In many cases, these kids wind up showing normal responses and measures. They're able to hear sounds, which suggests that their ears are largely functioning, and that it's other parts of the auditory system likely in the central auditory system, the parts of the brain that receive their connections from the auditory nerve, that may be a miss and could prohibit kids from tuning in to voices.

Nicholas Weiler (04:42):

So it's something further downstream, something in the brain about how the voices are being processed.

Dan Abrams (04:46):

That's our working hypothesis. And if peripheral hearing is considered to be relatively normal, then it would strongly suggest that it is a downstream difference that is causing kids with autism from tuning into the voices around them.

Nicholas Weiler (05:03):

And so is the idea that whatever the cause of the challenge with paying attention to voices, that then leads to a number of the other challenges or is tied up with other challenges in socialization that occur with people with autism?

Dan Abrams (05:18):

Yeah, we think that it plays a role in social development in kids with autism. The extent of that role is an important question, but the way that we think about it is that if kids aren't tuning into the vocal world around them, they miss out on lots of opportunities to learn about the social world.

(05:37):

We learn about how people express themselves, phonological learning. There's just countless lessons about the social world that we gain through vocal experiences, both observed vocal experiences and also ones that we participate in, child-directed vocal experiences. And so if a child's tuning out, they're missing out on lots of social learning.

(06:02):

I mean, we certainly want to kind of couch this in a much broader context, which is that there's so much to socialization and social learning and our social world, and I think we take for granted how important voices are and what voices are in our lives. We're so used to hearing them, and in many ways we're so used to so many aspects of the social world.

(06:25):

The two of us doing this back and forth thing where I talk and you listen and we're looking at each other. I know we're in two dimensions here, but we still know what to do. You talk, I listen, I talk, you listen. I have a sense of how you're feeling. You're reacting to my vocal tone, and we're just engaged in this back and forth that's actually complex and there's so much to it.

(06:47):

And we've learned so much about the social world to get us to this moment where we can do this effortlessly. I think it's easy to take for granted how much learning and how complex this process is. Also, on a side note, how much of our brain is required to do this thing?

(07:08):

We're looking at each other, we're listening to each other, we're understanding each other. I'm anticipating what you're thinking and what you might say next. It's just the social thing that we do requires vast neural resources.

Nicholas Weiler (07:23):

I mean, we're such social animals. Our brains are just wired for this stuff. But it takes experience and it takes learning. As you said, even babies are starting to pick up on this tennis match idea.

Dan Abrams (07:34):

Oh yeah, they know right away. I remember our first son's... We were in the hospital. He had just been born. He was literally, I don't know if he was even a day old, and we had been talking to him pretty much nonstop for that first whatever 18 hours we had been there.

(07:50):

And then someone else walked in the door. It was a nurse who had a very different way of speaking. And I'll never forget the look on our son's face when he saw this woman walk in and start speaking. He knew right away this was a totally different social experience and was wide-eyed and staring at her. Kids are just tuned into this stuff. And again, he has no idea what she was saying, but he was locked in and he was wired to attend to.

Nicholas Weiler (08:22):

And that seems like such a important foundation, as you were saying, for all the complexity that comes after. So one of the things you were saying about the kids who are on the autism spectrum is often it seems like a hearing problem, but turns out that there's something going on further downstream in the brain.

(08:41):

And we talked before about with the mother's voice and the children versus adolescence, there's this network of the hearing center, the voice processing center, which is then connected to reward, what makes voices so rewarding and salience, what makes them so interesting, and these other pieces of the brain that are involved in social cognition.

(09:03):

So it seems like the next question must be, what is different in this network in kids on the autism spectrum that makes them less likely to attune to people's voices from a very early age? What do you think is going on here?

Dan Abrams (09:16):

Yeah, great question. And this really strikes at the heart of what we're trying to answer. The most straightforward way of conceptualizing some of these brain regions is to think about what part of the brain is responsible for these kind of voice processing challenges in kids with autism.

(09:34):

On the one hand, you have what we might consider the "hearing brain." And the idea with the hearing brain is we have these auditory subcortical and cortical regions that are considered part of what we call the ascending auditory system. These are brain structures that are activated by all different sounds.

(09:53):

They receive connections from the eighth nerve, the auditory nerve, and they're highly responsive to sounds, and they're known to distinguish different acoustical features. Did I say boop or did I say bip or did you hear this tone or that tone, or was this sound longer than that sound, and these low level acoustical features.

(10:17):

So that's our hearing brain. We know all about the hearing brain, what the connections are, what the different nuclei in the ascending auditory system, all the way up to auditory cortex are responsive to and what acoustical features they care about. And then beyond the auditory system in the brain, we have this whole other kind of slew of brain systems that allow us to "make sense" of the sounds.

(10:43):

We process meaning from sounds, we have to contextualize sounds, we have to identify who is speaking. We have to do a whole bunch of stuff when we hear a voice. And we generally think that there are lots of other brain systems outside of our hearing brain that allow us to make sense of the sounds that we're hearing to identify who is speaking, etc, etc.

(11:12):

And so a fundamental question that our brain studies attempt to address is the reason kids with autism tune out from voices, is it because they're hearing brain has trouble making sense of the acoustical features in speech? Or alternatively, is it the case that other brain systems may present a bottleneck for making use of vocal sounds? I guess at some level, these two things, it's these aren't necessarily mutually exclusive hypotheses.

(11:43):

I mean, imagine if the auditory system, if the kind of hearing brain was struggling when you hear voices, I man, the way I think about that is maybe voices sound kind of noisy like a staticky AM radio to a kid with autism. Hard to pick out, maybe not pleasant to listen to. And so there may be something at the hearing brain level that makes voices a noisy signal that's difficult to discriminate, and that could potentially lead to weaker connections and communication with other brain systems that can process reward or salience or other things.

(12:23):

So this kind of sensory model may not be mutually exclusive from thinking about other brain systems that may struggle to make sense of those sounds. So again, it's a little bit messy here.

Nicholas Weiler (12:37):

I mean, it seems like that also would line up with, I understand that a common struggle for people with autism is this sensory overwhelm. That there's certain sounds, certain feelings, that are just very aversive, very unpleasant. And so you could imagine that, well, we know that there is some sensory stuff going on for lots of kids with autism, and so it makes sense that that would likely affect voices as well.

Dan Abrams (13:02):

Do you mind if I jump in?

Nicholas Weiler (13:03):

Oh yes, please.

Dan Abrams (13:03):

That's a very astute point, Nick. Up to 90% of people with autism have some kind of sensory processing difference or difficulty. It's a hallmark of autism. It could be somatosensory, could be vision, it could be auditory, it could be all of the above. Sensory level processing differences are extremely common in autism.

(13:25):

And so it makes sense that here's this auditory signal. There may be some sensory processing challenges associated with voices. And I don't think there's any question that people with autism do have some different sensory experiences as they relate to voice processing. I guess the question that we think is important is are those the kind of processing difficulties that lead to vocal engagement issues that we see in kids with autism?

(13:55):

The reason I think that's important is I don't think there's much question that the sensory experiences of kids and people with autism is quite different from neurotypicals. But I'm also, I think it's a reasonable question to ask, is that the reason that they're tuning out from voices? That's an important question in the context of these sensory theories. And one, it's a question that hasn't been answered and that we hope to address in our work.

Nicholas Weiler (14:23):

And so one of the questions would be, and I'm curious if there's evidence here, do we have any suggestion about whether the actual auditory processing, how kids hear the voice is affected versus whether that is a rewarding or pleasant stimulus, whether that activates these other socially important brain regions that we've been talking about? Do we have any way at the moment of distinguishing those or does that require some brand new science?

Dan Abrams (14:49):

No. So I mean, work using fMRI in kids with autism has addressed these exact questions. And we think that if sensory level hearing brain related impairments were a major contributor to vocal challenges in kids with autism, we would see a very specific neural profile in these kids. We know where the hearing brain is, we know roughly what the hearing brain does.

(15:18):

And if auditory processing were a significant bottleneck for discriminating voices, we would see a very specific profile of neural activity and brain connectivity as measured with fMRI. And alternatively, if we think that auditory system is largely intacting kids with autism and that its downstream systems, other systems associated with salience, reward, social evaluative processing, if we think that those systems may be more closely related to voice processing difficulties in autism, we would hypothesize a different neural profile in kids with autism.

(16:01):

So all of our studies test these hypotheses of what these brain systems do to help us make inferences about what may underlie these voice processing difficulties in kids with autism, because it's very difficult to just use behavior and truly understand why kids with autism are tuning out from voices.

Nicholas Weiler (16:22):

There could be so many reasons.

Dan Abrams (16:24):

There could be lots of reasons. And the way we think about our work is we're trying to use the brain to gain some additional insight into why this may be. If it's a sensory level processing difficulty, we would hypothesize these auditory processing regions. And if it were a different kind of reward salience processing issue, we would hypothesize a totally different collection of brain regions.

Nicholas Weiler (16:49):

So you were telling me earlier as we were talking about recording this interview, that there's this developing idea that you and colleagues have been thinking about that maybe that there's this sort of hub network that one of the main things that's different in the brains of kids with autism. Can you give us a sense of what does that hypothesis look like and why do you think that might be an important way of describing what's going on in the autistic brain?

Dan Abrams (17:14):

Yeah, so as I mentioned, we have these auditory structures and their in superior temporal cortex and include a series of tightly interconnected brain regions in temporal cortex. But ultimately, these auditory processing regions have to at some level connect to other brain regions so we can make sense of the vocal world around us. And a question as well, what part of the brain allows us to connect these hearing parts of the brain to these other systems that help us identify different voices that help us make meaning from speech sounds?

(17:49):

And we think that a key part of the brain for this is called the superior temporal sulcus. And this part of the brain in neurotypicals has really extensive, much greater connectivity to prefrontal and parietal brain systems compared to all other parts of superior temporal cortex.

Nicholas Weiler (18:11):

And just to unpack that anatomy a little bit for listeners, so this superior temporal sulcus, that's in or close to the auditory, the sort of vocal processing regions in the brain?

Dan Abrams (18:21):

Correct, yeah.

Nicholas Weiler (18:22):

And so in case anyone's interested, superior means above, temporal means at your temples, so right in front of your ears. And then sulcus is the little in foldings, in our brains we've got this folded thing. So the sulcus means on the inner fold there.

Dan Abrams (18:39):

That's a perfect description.

Nicholas Weiler (18:40):

And so then you're saying you look at the auditory, the vocal areas, and this particular fold of the brain is much more connected to parts of the brain that you would expect to be involved in reward processing or in social evaluation, things like that. So that gives you this hint that like, oh, maybe this is the thing that's connecting how we hear the voice to what we make of it.

Dan Abrams (19:05):

Yes. And we weren't the first to think of this possibility. Others have written about this, but our study was among the first to show how tightly connected the STS was to these other brain systems. And so here's this notion, and one of the takeaways from that study is one can conceptualize the STS as being a hub.

(19:26):

It provides a critical connection to link different kinds of brain systems, both auditory systems that are associated with hearing sounds and superior temporal cortex and linking these to all kinds of other brain systems outside of temporal cortex. That's an important concept. Here, we have this evidence of a hub in the STS and then linking it back to our autism research.

(19:53):

What we found in several studies is that one of the common denominators among our voice processing brain research in kids with autism is disrupted functional connectivity between the STS and different brain systems associated with social evaluative processing, reward processing, and salience processing.

(20:18):

And so listen, our autism research is, as you might imagine, very difficult. We don't always wind up with the sample sizes we wish we had. It takes years to collect these data. It's often quite frustrating how dismissive reviewers are when they only see that we had 22 kids in our study. But for example, a team will work tirelessly to collect those data over several years, just to get 22 kids with autism to keep still in the scanner long enough to have enough data to interpret.

(20:50):

And so just like anything in neuroscience, we look for patterns. Seeing something in one study is not generally enough to convince many of us, but to see repeated patterns across studies is an important barometer and may begin to convince people that this is a possibility. [inaudible 00:21:13] people are very conservative and they should be. We're talking about a very important population.

(21:19):

These results affect families and parents. And so I think a hefty amount of conservative viewing of these data is appropriate. And that's why I say we've seen these kind of patterns across several studies that appear to link to this dysfunction of this STS hub. One of the contributing factors, and again, I want to be very careful about how I phrase this.

(21:44):

I don't want to say this explains everything about voice processing and autism. I certainly don't want to say this explains everything about social function and autism based on our work and others. It appears that a contributing factor to voice processing difficulties is this dysfunction of the STS in how it is able to communicate and share information between auditory processing regions and superior temporal cortex and these other downstream brain systems associated with reward and social evaluation.

Nicholas Weiler (22:17):

So what it sounds like to me, and correct me if I'm misunderstanding this, you're saying, yes, there may be issues going on with vocal processing, with sensory issues. We know that there are sensory involvement in autism, but another important thing that we need to be considering is that even after the vocal processing, parsing the voice, there seems to be this pattern of dysfunctional or lower levels of connectivity between hearing the voice and the brain regions that would tell you that this matters, that you should pay attention to this, that this is important.

(22:51):

So two questions from that. One, would that be a change in the way the field has been thinking about this? And two, if so, what does this imply about how we should think about helping kids work through some of these challenges so that they can be better integrated into society?

Dan Abrams (23:10):

So I agree with your assessment of this, which is we don't know how much sensory contributions relate to these other downstream differences that we're seeing. And I'll be honest, the idea of a social reward difference and deficit in kids with autism, individuals with autism, has been around for a long time now. What was new about this work is really there wasn't a ton of research about voice processing in the brain in autism.

(23:39):

There's much more studies of faces. Visual processing was the basis for the original social motivation hypotheses. And so we know certainly much more about vision and autism than we do about hearing, but auditory processing of voices is a critical signal to help us be connected. Hearing voices instantly of our loved ones instantly connects us with them, provides social belonging and all kinds of important social information. So these ideas, many of these ideas have been kind of tossed around before, but not in the context of hearing voices, which is one of the novel things about our work.

(24:21):

So some of the things that we think about, and for what this may say about interventions for individuals with autism, the way that we think about it is that an important direction for having kids tune into the vocal world around them may be... Here, our research has implicated salience and reward processing parts of the brain, and separately, over the last 30 or 40 years, there's been clinical effort focused on interventions that naturally motivate kids with autism to engage in verbal communication.

(25:01):

They follow under an umbrella called NDBI, Naturalistic Developmental Behavioral Interventions, and that's an umbrella term for several intervention approaches that really focus on finding ways to naturally motivate kids with autism to engage socially and to engage in their everyday life away from a clinic, a sterile clinic with flashcards or doing rote memory tasks.

(25:30):

But to take the intervention into their everyday lives, focusing on things that they want to do, whatever their particular interest or interests are. One of my awesome colleagues, Lynn Cagle and her husband, Bob, 30 or 40 years ago, they came up with an intervention called Pivotal Response Treatment, which is one of the foremost and most popular of these NDBIs, Natural Developmental Behavioral Interventions. And they've had great success with it.

(25:59):

They teach the parents how to deliver this intervention in their everyday lives with the child, how to take their interests and turn them into opportunities to generate and to improve social function. And so here are these kind of brain results in the one hand that have identified social reward as a key contributing factor. And then on the other hand, we have these clinical interventions that are associated with improving reward, driving and engaging kids and trying to naturally motivate kids with autism to engage.

(26:34):

And so from one perspective, the implication of our work is that I think it provides some meaningful neuroscientific context for thinking about strengthening social reward, and that a key approach for building social skills in kids with autism is to find ways to make socializing more rewarding for them.

Nicholas Weiler (26:58):

Well, if we have neuroscience that supports the idea that there's some sort of disconnect between hearing voices and the rewarding nature of social connection that gets to the fundamentals of how we think about autism, fundamentally, there's a disconnect and the social world that neurotypicals take for granted as being rewarding.

(27:23):

We love chatting. We love this social engagement. We love talking and listening to people's voices and watching movies and listening to podcasts. If that connection is not there, then maybe we know the brain is plastic, particularly early in life. And so maybe finding ways to, as you say, connect some of those valuable social skills or foundations to things that kids do care about, the things that are rewarding, could be a really nice way of helping to get them some of the skills, get them some of the ability to integrate that would be helpful to them.

(27:56):

I know that we're running a little short on time here. Is there anything else you'd want listeners, particularly people maybe who have kids with autism in their families to take away from this conversation?

Dan Abrams (28:05):

Oh, I guess I wanted to, just on that last point, I guess I wanted to just make a quick little statement that actually one of our ongoing studies in collaboration with Dr. Lynn Cagle and Bob Cagle investigates whether improvements in social function following PRT drive neural plasticity in voice and reward processing brain regions in kids with autism. We basically bring kids in, we have them listen to our vocal fMRI tasks, and then they do a 12-week intervention, and then we have them come back after the intervention and do more voice processing tasks using fMRI.

(28:46):

And the question is, do kids with autism who show improvement from the intervention, do they also show strengthening of some of these exact kind of brain circuits that we've been talking about? And our hypothesis that there is a relationship, and so one of the questions we're asking in the next year, we'll see if those hypotheses what winds up happening in the brain following improvement from PRT.

Nicholas Weiler (29:11):

Fantastic. Well, I'll definitely, if you've got any open trials, we'll put some of those links in the show notes, and if listeners know anyone who might be interested, well, maybe we can connect you up.

(29:22):

Well, Dan, thank you so much for joining us on the show. This was really interesting and gave me a very different perspective on what we understand about what's going on in the autistic brain.

Dan Abrams (29:30):

Well, thanks so much for having me. It's a pleasure. I guess if I have one kind of last message for parents.

Nicholas Weiler (29:36):

Sure.

Dan Abrams (29:36):

I always want to be cautious about our brain findings, because they're in relatively small samples and we try to replicate as much as we can and all these things, and I guess I don't want to make it sound like we have all the answers. The work that we do to understand the brain in kids with autism is imperfect. We're using the best tools that we have available to us.

(30:01):

It's very difficult work, and it takes a long time and real investment to start to get answers about brain function in kids with autism. This is a long-term challenge, and I hope that as a field we'll be able to provide more clarity about brain function and refine our hypotheses. I hope everyone knows that this is mid-process here for truly understanding these challenges in kids with autism and how they relate to brain function.

Nicholas Weiler (30:32):

I really appreciate that. It's important to remember that science is an ongoing process. We're always learning. We're always improving our understanding and that everything we say is going to be provisional, and so we just have to work from that and keep investing in this basic understanding of the brain and how it connects. Dan, thank you so much for joining us on the show. This has been a really interesting conversation.

Dan Abrams (30:55):

Thanks so much for having me.

Nicholas Weiler (30:59):

Thanks again so much to our guest, Dan Abrams. He's a clinical associate professor in the Department of Psychiatry and Behavioral Sciences at Stanford Medicine. To read more about his work, check out the links in the show notes.

(31:11):

If you're enjoying the show, please do subscribe and share it with your friends. This is how we grow as a show and bring more listeners to the frontiers of neuroscience. We'd also love to hear what you think of the show, what's working for you, what's not working for you. It really makes a big difference to us to hear from listeners, and so we'd love for you to contact us. Share a comment on your favorite podcast platform, or send us an email at neuronspodcast@stanford.edu.

(31:36):

From Our Neurons to Yours is produced by Michael Osborne at 14th Street Studios with sound design by Morgan Honecker. I'm Nicholas Weiler. Until next time.