Autism Spectrum Disorder: Behind the Scenes

In this episode, we go inside the brain itself, using a building as our guide.

We talk about what early altered brain development actually means at the cellular level, and why the blueprint gets drawn differently before anyone notices. 

Why the structures most responsible for threat detection and emotional memory are built larger in childhood, and what happens to them after that. 

What the amygdala and hippocampus are actually doing in a crowded, unpredictable environment, and why that costs so much. 

Why there are still no reliable biomarkers for autism, and what the structural findings tell us about why that's not a simple problem to solve. 

What neuroinflammation looks like inside donated brain tissue, and why the maintenance crew never stood down. 

And what happens when you put these two papers side by side, because they don't just describe the same condition from different angles. They answer each other's questions.

All of it from one paper. 

Shuid, A.N. et al., Update on Atypicalities of Central Nervous System in Autism Spectrum Disorder. Brain Sciences, 2020.

Transcript

SPEAKER_00 0:00

Think about the last time you walked into a room and felt unbelievably overwhelmed. Maybe it was a party, or maybe it was the shopping mall on a weekend, or maybe it was the school cafeteria at lunch. Think about it. You're surrounded by hundreds of kids, a hundred conversations. Chairs scrapping, trays clattering, and the smell of everything mixed together all at once. For most people, those environments are just loud. Annoying at most, but mostly manageable. You tune it out and you move on. But for a lot of people on the spectrum, that same environment doesn't register as background noise. It registers as a signal, all of it, at once, at full volume. Not necessarily noise in the room, but noise in the nervous system. And when someone tries to explain that, or when, you know, if you've noticed your child comes home from school and needs two hours of pure silence, or an adult declines an invitation to an event that they really, really, really wanted to go to, but they couldn't explain the cost of exhaustion that would come afterwards if they did go. The response they're often faced with is but nothing serious happened. You were just at a party. What today's episode is about is the gap between those two descriptions, because there is a biological explanation for that gap. Not one explanation, but several. And they start inside the brain itself, in the way it was built, and in the way it has been running ever since. In episode one, we laid the foundation. We built the airport with the help of the Lancet Review from 2018. We discussed what autism is, how it's diagnosed, the genetics, the outcomes. That was all the broad clinical picture. Today, I want us to go inside the building, behind the walls, into the structure itself and the wiring that runs through it. We're working from a 2020 review in Brain Sciences by Shued and colleagues, a cellular and molecular look at what is different about the central nervous system in autism. Since the paper I've picked is quite dense, I decided to split it over two episodes. So today's episode might be a little bit shorter. We're going to cover two mechanisms, brain structure and neuroinflammation. And today I want to do something the first episode started, but we couldn't quite get into. I want to show you what happens when you put these two papers side by side. Because when you do, they don't just describe the same condition from different angles, but they answer each other's questions. This is a show about the actual science, not the headline, not noise. We take one paper, we read it carefully, and we build something you can hold on to. I'm your host, Nafisa. I'm not a clinician, I'm someone who believes that good research deserves a wider audience and that most people, given clear and honest information, are capable of doing something meaningful with it. Today is episode two. We're still in the same arc discussing autism. Still in the airport, but today we're going behind the scenes. Decisions get made about the size of the room, how they connect to each other, where the corridors go, what happens when you move from one space to the next. In typical brain development, the brain builds itself according to a set of biological instructions. Neurons form. They migrate to where they need to be, they connect, they get refined. The blueprint isn't drawn by anyone in particular. It's carried in the biology. But in autism, some of these instructions run differently. And the resulting architecture, the size of certain rooms, the connections between them, doesn't always follow the expected plan. The same way blueprints are drawn before a building is built, ASD brains, based on the research, are also always the same way before they get diagnosed. Autism doesn't develop at three when they get diagnosed. The Lorde review was clear about this. ASD results from early altered brain development. The biological differences are there before the behavioral signs become visible, like the blueprint of the airport. But what Lorde couldn't do in a clinical overview is show you which structures are different and what that means. That's what the Shued Review does. It takes the phrase early altered brain development and goes inside it. Two structures come up more consistently than almost any others in brain imaging in autism research. The amygdala and the hippocampus. Let me tell you what they actually do before I tell you what the research shows, because the names are meaningless without the function. The amygdala is a small almond-shaped cluster of neurons buried deep in the brain. You have one on each side of your brain. Think of it as the building's threat detection system. Its job is processing emotional intensity. So questions like: is this safe? Should I pay attention to this? Does this require a response? It operates in fractions of a second, mostly below conscious thought. You can usually sense it in a social environment, maybe a classroom or an airport or a party. It's working constantly in the background. Every face you scan, every unpredicted noise, every shift in the room is being evaluated by this system in real time. And it's very expensive to run, neurologically speaking, of course. And it doesn't get a break when the environments keep changing. It has to adapt according to its environments. The hippocampus sits right alongside it. Its job is memory and emotional context, essentially attaching meaning to experiences. Think about it as the archive room, the place where experiences get stored, patterns get learned, and contexts get attached to everything the brain encounters. It's the reason why you already know a particular environment might be overwhelming before you even walk into it. Because there might be learned context from the last time you did this. It connects the amygdala's present moment threat signal to everything the brain has already learned. So think about it this way: the amygdala takes in information and then the hippocampus understands the context that information was taken into and it stores it together. These two structures are in constant conversation. So when we look at what's different about these structures in autism, we're looking at systems that are central to social perception, emotional regulation, and threat response. All of the things that make a busy, unpredictable environment either managing or super overwhelming. So let's get into what the research shows. Across multiple MRI studies, brains with ASD consistently showed enlargement of both structures in childhood, and in some cases significantly more than others. Although findings vary across age groups and studies. One study measured brain structure in infants between 23 and 27 months and found significant enlargement in the ASD group, and that amygdala enlargement in early childhood was directly linked to the severity of social communication and emotional difficulties. So essentially, the bigger the threat detection system early on, the more pronounced the social communication and emotional challenges. Here's the first place where I want these two papers to talk to each other. Lorde et al. described a pattern of overgrowth of brain volume in infancy and early childhood as one of the most replicated findings in autism neuroscience. But Lorde couldn't tell you where in the brain that growth was concentrated or what it meant for the person living with it. The Shute findings answer that. The growth isn't uniform. It concentrates in structures most responsible for threat detection and emotional memory. A brain building those systems larger during the years when every experience is new and the brain is most actively wiring itself up is a brain that from very early on is calibrated to treat its environment as more demanding than a typical brain would. But here's where it gets complicated and more honest. Studies in adults with autism show the opposite, reduced amygdala volume compared to controls. Larger in childhood, smaller in adulthood? That sounds contradictory. It's actually developmental. The amygdala in ASD appears to be overbuilt early, and then it doesn't undergo the normal age-related growth that typically developing children experience. Essentially, it's like saying that you wanted to expand the operation services in the airport, but the entire expansion is stalled. This is the second place where the two papers connect and where Shute answers something Lorde raised but couldn't fully address. Lorde was explicit, there are no reliable biomarkers for autism, no blood tests, no brain scan, results that you can point to. Diagnosis has been made on the basis of behavior. And most people ask, why? If the differences are biological, then why can't we measure them? The amygdala data is part of that answer. The same structure gives you different readings depending on when you measure it, enlarged at two years old, but reduced by adulthood. A biomarker that needs to be taken in exactly the right developmental window in a condition that isn't diagnosed until behavior becomes visible, which is often years later, isn't a reliable biomarker. It's a moving target. There is one more finding from this section. A study looked at two adults with bilateral amygdala damage, which means that both sides had been significantly lesioned. Researchers assessed them thoroughly for ASD features. Neither showed symptomology, despite the damage. What this tells us it's not the amygdala itself that's the problem, it's the connections, how the amygdala talks to the rest of the brain. A room can be the wrong size and still function. What breaks the system is when the corridors between the rooms don't carry information properly. Lorde described this pattern as underconnectivity between distinct brain regions. Parts of the brain that should be coordinating aren't doing so effectively. The amygdala lesion study in Shude may explain that from a completely different direction, two independent lines of evidence arriving at the same conclusion. In autism, the corridors between rooms matter more than the size of any individual room itself. Here's the cross-analysis I think is most clinically meaningful in this section. Lorde documented anxiety as one of the most common co-occurring conditions in autism. Social anxiety, generalized anxiety, anxiety increasing in adolescents, particularly in girls. Lorde presented it as something that travels alongside autism. But the shoot structural findings suggest it may not just be traveling alongside. It may be mechanistically connected. This is a threat detection system that was built larger, calibrated to treat the environment as more demanding and wired differently into the rest of the brain from early development. That's a system already predisposed to find more things threatening, not because of excessive worry, but because of architecture. This doesn't mean anxiety and autism is untreatable, but it does mean that treating it as a purely psychological problem without understanding the biological substrate it's running on may be part of why some approaches work less well than expected. The blueprint was drawn differently, and the differences matter most, not in any single room, but in how the rooms connect or communicate to each other and whether the building can coordinate itself effectively when things get demanding. Okay, let's say that the blueprint was drawn and the building went up. Now let's talk about the electrical wiring. In any complex building, the wiring is invisible to almost everyone who uses it. It runs behind the walls, it carries information from one system to another. When it's working, you don't notice it. But when it overheats, when the current is too high or the system is running something it wasn't designed for, things start to go wrong in ways that are hard to trace back to a single source. That's the best analogy I have for what neuroinflammation looks like in the brain. And it's one of the most consistent findings in this paper. There is something a lot of people with ASD describe, and that researchers have started to take seriously, that for a long time got dismissed. The feeling that your body is working harder than it should be, that simply existing in a classroom, an office, a social environment has costs that other people just don't seem to pay. That they can spend an entire day doing what looks from the outside like perfectly ordinary things and come home feeling like they've run 10 marathons. We built one part of that explanation in section one. A threat detection system at a high setting from the beginning. But there's another layer underneath, a cellular one, and it's called neuroinflammation. I want to explain this carefully because the word inflammation carries a lot of associations. Like most of them are physical, visible, and acute. For example, a swollen ankle, a fever. It's quiet, it's persistent with effects that accumulate over time. To understand it, I need to introduce two cell types. You don't need to know the names. What matters is what they do. The first one is microglia. Microglia are the brain's own immune cells. They're found throughout the brain and spinal cord, and their job is surveillance. Think of them as the building's maintenance crew, small, mobile, constantly scanning. Under normal conditions, they move through the brain continuously, clearing debris, supporting neurons, keeping things running. They're in maintenance mode, not emergency mode. Just like a building's janitorial team, present, necessary, invisible when everything's working. But when something disrupts the normal environment, an injury, an infection, something goes wrong in how neurons are functioning, microglia activate. When they activate, their behavior changes. They switch to emergency mode. They release inflammatory signaling proteins called cytokins that are described to coordinate an immune response. Under normal circumstances, this is exactly what you want. Active, address the problem, stand down. The question the research asks in autism is what happens when they don't stand down? The second cell type is astrocytes, sometimes called astroglia. Think of them as caretakers. The people who make sure the building is clean and the temperature is right for everything else to function well. Their job is to maintain the neurons that live in the environment, regulating chemical balance, clearing excess signaling molecules after they've done their job. Under normal conditions, astrocytes are supportive and regulatory. But when they get activated in response to injury or sustain neural dysfunction, they can start releasing the same kind of pro-inflammatory proteins as microglia. Instead of caretaking, they start contributing to the alarm. Now let's get into what the research actually shows. Think about what happens when a building's emergency system gets triggered. The fire alarm goes off, the sprinklers activate, the maintenance crew drops everything and switches into emergency mode. All of it is designed to contain damage and protect the structure. But now imagine the alarm never stops. Not because there's still a fire, but because something in the system got stuck. The crew is still running emergency protocols, but the sprinklers are still on. The building is spending an enormous amount of resources managing a response that was never designed to run indefinitely. That's the closest analogy I have for what the postmortem evidence shows in a brain with autism. Postmortem research means exactly what it sounds like: studies that use brain tissue donated after death. It's some of the most direct evidence we have about what's happening inside the brain itself, rather than measuring blood or behavior from the outside. In one study by Vargas and colleagues, researchers measured dozens of inflammatory proteins in donated brain tissue from individuals with ASD and compared them to a control group. What they found consistently across multiple brain regions was that the alarm was still on, which means that there were elevated inflammatory markers. And when they examined the tissue directly and they looked at the actual cells, they found marked activation of both microglia and astroglia. The maintenance crew and the caretakers, both in emergency mode in the frontal cortex and the cerebellum. A second study by Lee and colleagues found the same thing in the cerebral cortex. Elevated inflammatory proteins, a level of immune activation the researchers described as indicative of heightened immune response, potentially associated with localized brain inflammation. Now, the frontal cortex and the cerebellum aren't obscure backrooms. These are the regions you run hardest when you're in demanding environments. The frontal cortex takes care of self-regulation, decision making, the management of competing demands. The cerebellum handles timing, coordination, and sensory integration. In an airport, in a noisy office, in a school cafeteria, in any space where a lot is happening all at once, these are the systems being asked to do the most work. And in the brain tissue of individuals with ASD, those are the exact rooms where the emergency lights are always on. What does that mean in practice? Resources spent running an internal alarm that isn't available for anything else. For filtering sensory input, for reading the room, for regulating the response to everything that's happening around you, the building is diverting power to a maintenance response, and that power is coming from somewhere. And here's the part that matters most developmentally. If that activation was happening during early development, before age three, when the brain is at its most intensive building phase, when the wiring is literally being laid down, it doesn't just cost energy. It can interfere with the construction itself. Chronic microglial activation can disrupt neurogenesis, which is the birth of new neurons, and can affect how they migrate and form connections in the brain. The wiring for the emergency lights gets laid in the building before the walls go up. Here's where the two papers connect again, and where shoot starts to fill in what something Lorde left open. Lorde's review puts autism heritability at 74 to 93%. Very high. But Lorde also listed environmental risk factors associated with increased risk in autism. Some of it was maternal infection during pregnancy, valparic acid exposure, preterm birth, maternal immune conditions. And Lorde was honest that the mechanism by which those environmental exposures translate into altered brain development wasn't fully understood. The neuroinflammation findings are a plausible answer to that question. Maternal infection activates the immune system, and that activation can cross into the fetal brain and trigger microglial activation in a developing nervous system that's still building itself. Preterm birth creates the kind of biological stress that primes an inflammatory response. In a brain that is already genetically predisposed to certain vulnerabilities, an environmental trigger that activates inflammatory pathways during the most critical window of development could be exactly how those risk factors translate into altered wiring. The genetics set up the building, the environments affect how the wiring runs during construction. Lord found the associations. One more connection before I close this section. Preview of where we're going next. The Lourdes Review noted that approximately 47% of people with autism experience gastrointestinal symptoms. Nearly one in two. It's one of those findings that sits in the clinical literature at a well-documented fact without a clear explanation of why. The neuroinflammation findings are the beginning of that explanation. Because here's the thing about wiring in a building: it doesn't stay in one room. The immune system is not contained to one location. The same inflammatory process that operates in the brain can operate in the gut. The gut has its own nervous system, its own immune tissues, and its own relationship to the central nervous system. A body where neuroinflammation is chronically elevated may have parallel processes running in places that look unrelated, but are. That thread gets its own full episode. But it starts here, in the wiring. Before I move on, I want to name the limitation the paper was honest about. A lot of the early immune research in autism measured blood, not brain tissue. The findings are real, but blood isn't brain, and what's elevated in the bloodstream doesn't always get reflected in what's happening inside the central nervous system. And the postmodern studies that do look directly at brain tissue used small samples, eight or eleven people. Meaningful findings, but they still need larger replication. The paper doesn't hide that. And neither do we. Okay, let's bring this all together. We came in with a question: why does a normal day cost so much from the inside? The structural findings gave us one answer. The brain regions most responsible for detecting and responding to social and emotional information were built larger and connected differently. A threat detection system turned up high before the traveler even arrived. That's not a character trait, that's architecture. The neuroinflammation findings gave us another. A brain where immune response may have been chronically activated, possibly from early development, possibly triggered by the same environmental exposures, Lorde said, is a brain spending resources on an internal process that then isn't available for anything else. And when you hold these two papers side by side, things become clearer. Lorde documented anxiety as a co-occurring condition. Shuws you the structure it may be running on. Lorde found gastrointestinal symptoms in nearly half of the people with autism. Shu chose the immune process that could connect the brain to the gut. Lorde said no reliable biomarkers exist. Shu shows you exactly why, because the same structure looks different depending on when you measure it. These are two separate papers about autism. In fact, they're two papers in active conversation with each other, and the conversation isn't finished. If you're a parent who has tried to explain why your child comes home exhausted from a day that looks fine, this is the biology of what you were describing. The building was constructed differently. The wiring has been running hot. Neither of these things are your child's fault, and they're not yours either. If you're someone with ASD who has spent years being told you should be able to manage what everyone else seems to manage, this research is part of what you've been trying to say. The cost is real. It's biological, it's not an attitude. Next episode, we go deeper into the chemical communication system, the signals that travel between neurons, and the energy production machinery inside brain cells themselves, into why processing a demanding environment may be genuinely more expensive at the cellular level for most people with ASD, and what that means for the kind of tired that makes no sense to anyone watching from the outside. If this episode gave you something, a language or clearer picture, something you've been looking for, share it with one person who needs it. That's how this research reaches people who wouldn't otherwise find it. And remember, outcomes change when environments change. That's not optimism. That is what the data shows. Things can be redesigned. They just have to be built brick by brick.