Regina Sullivan remarks

Unlearning Violence: Evidence and Policies for Early Childhood Development and Peace

February 14, 2014
ASEAN Auditorium
The Fletcher School
Tufts University


Transcript of remarks by Regina Sullivan

Thank you very much. It’s really an enormous pleasure to be here and talk about some basic neuroscience to you and why I think it’s important to understand brain development as one thinks about the effects of trauma throughout the lifespan and its enduring effects.

I think people typically think of the brain of a child as simply an immature version of the adult brain. So brain areas come online and then switch on some adult activity. For example, the prefrontal cortex comes on line in adolescence and then the child is all of a sudden able to have prefrontal cortex activities that we have. So it really is important to understand that the young brain is different and understanding how trauma in early life produces vulnerability to later life violence lies in our understanding the unique functioning of the child’s brain. The child’s brain is not an immature version of the adult brain and, of course, all of you know that because if you just think of feeding in the infant, the infant suckles. It doesn’t eat the way we do: the perfect example how the brain is organized for the infant.

The other thing that’s really important is that the infant is designed to attach to their caregiver. Things that happen to the child happen within a context of a brain designed to make sure the caregiver takes care of them. So when we think about trauma, we have to also think about how the attachment system plays a role. The infant brain is structured for unique challenges of early life, attachment, and acquiring nurturing from the caregiver are critically important.

So if the brain is designed for attachment, what does that mean? I love to show this picture of imprinting - Konrad Lorenz showed this in the 50’s. When these baby birds hatch, they attach to the first moving object. And whatever that moving object is, the attachment learning system is activated and it produces very specific behaviors related to proximity seeking. I’m going to argue that the same thing happens in rodents, which I’ll talk about today.  In humans there are certain things that have to happen in the environment for the human baby to attach to their caregiver, and it’s important to note which things are or are not important.

Do we really need the entire human to get the attachment or is it certain behavior that the caregiver does to make that attachment take place? Look at this attachment formation. It almost looks as though it’s genetically determined and then at birth the attachment circuitry just occurs, and that’s not the case. And you can look at this picture of chicks following Konrad Lorenz and it really illustrates that the attachment process is not just a predetermined event but there’s learning involved and the infant has to learn who the caregiver is, to show the attachment behavior.  During this process they’re also learning about the world via the caregiver behavior to the infant.

So when we think about the brain being designed for attachment, there are two basic things we have to remember. One is that infants learn to identify the caregiver but once learned, that attachment engages a biologically predisposed system supporting infant caregiver proximity. And secondly, during this process of attachment, there’s enduring effects because of brain programming by the caregiver. I’m also going to argue that during this process the child or the infant is also learning a lot and so some of the effects are going to be via learning. And I’m particularly interested in the effects that are caused by learning because I feel that they are more malleable for us for intervention.

So since this isn’t a neuroscience group, I just wanted to show one slide to illustrate the enormous change that occurs in the brain from birth to later life. And it’s not just the size of the infant brain is smaller:  it’s that cells are growing.  Here’s just one particular part of the cortex that is actually growing: huge numbers of cells grow in complexity over the course of a lifetime with experience and genes working together to program the shape of these neurons and how they connect to other neurons. That’s what’s happening during early life experience. So you’re actually changing the structure of neurons. Evidence shows that you can change some of these structural changes, sometimes you can’t.

So the genes and environment control brain activity throughout life. But early life experience also controls the developing structures of the brain. And here’s a slide I borrowed from somebody in the audience, showing that different brain areas come on at different ages during development. And this is illustrating synapse formation - that’s the connection between neurons. Brain areas for different functioning develop at different ages. The experience that a child has at this age, when vision is developing, is going to impact vision, if the experience occurs later in life, when the prefrontal cortex is developing then the experience will impact higher cognitive function. So when we think about trauma, we need to think not only about the attachment circuitry but at what age it’s happening because that’s the age and which brain area is developing.

In an ideal world, this caregiver relationship should be defining the experience base changes in the brain. Unfortunately, life is not always ideal. In my lab, we focus on the neurobiology of abusive attachments, but what we’ve learned about trauma can be applied to trauma outside of attachment and why it’s so important that the caregiver’s response to the child’s response to the trauma has such strong impacts on outcome. And I’ll try to give you some basic neuroscience data to explain mechanisms. I don’t want this to sound as though we have all the answer or I am going to tell you about the many things that can happen during development. Many, many people are working on this problem and it’s fairly difficult to identify mechanisms. So this is just a small piece of a very complex picture.

So it’s really great that this attachment system is so robust and that no matter what happens to a child, they attach to their caregiver. Even a severely abused child loves their caregiver. Social workers can try to rescue the child from this horrible situation but the child longs to go back to their caregiver. Their heart is broken from being separated from their abusive caregiver. And I will venture to say that attachment occurs throughout the lifespan and there are certain aspects and qualities of the attachment that remain constant throughout life. So, for example, a women can also form bonds and maintain bonds with an abusive spouse. It’s very difficult to get her to separate and when you are successful, the women or child many times wants to go back to the attachment figure but there are long-term costs.

I’m going to show you just a few neuroscience data slides to quickly illustrate some neural mechanisms for infant attachment to an abusive caregiver. We know from the literature that this early life abuse, even though it produces attachment, has later-life costs. There’s vulnerability to psychiatric disorders. There’s disruption to brain development. I focused on the amygdala, a brain area associated with emotion and emotional learning. In our research we ask how is the early life abusive attachment programing the brain and what is the child learning about the caregiver and how does this influence what the child learns about the world - is the world safe? When one talks about the child soldier, the attachment formation might be forming in that situation and I just think it’s something to think about as we think about trauma within that context and attachment.

In our research, we use a fear conditioning paradigm, and it’s a very simple procedure. In this room, if I was to do a fear conditioning experiment, I would maybe put some mild electric shock in your chair and have an odor come in just before the shock. You would learn two things from that. One is that this room is not a good place, and that’s called context learning, and it uses a brain area called the hippocampus. Children don’t have that in very early life, which means that they’re not learning about how bad the room is. And the second thing you would be learning is that the odor that I presented just before the shock predicts the shock. The weird thing about early life is that the brain area that supports that learning in us and in adult rats, the amygdala, is not online yet – it is not functional yet because it is still immature. So when infants have a bad experience such as abuse from the caregiver, it is not possible to form fear to the caregiver because the amygdala is not functioning. Instead this conditioning activates the attachment circuit.  We have lots of data from our lab that this occurs in the rat. It is not possible to assess the amygdala in children but behaviorally, children do form attachment to abusive caregiver.

Here’s just a very quick example of the attachment learning brain areas - the locus coeruleus, the olfactory cortex and the prefrontal cortex are all involved in the attachment circuitry and the primary neurotransmitter is norepinephrine. At about 10 days old in the rat, the amygdala - the brain center for fear - becomes functional, and the animal can learn fear. But the most amazing thing is that if the attachment figure is present, she actually suppresses learning within the amygdala and the fear circuitry. This means that in the older infant animal, where they should be able to learn fear, this fear learning is suppressed and the attachment circuitry is used. Thus, we have suppression of the adult version of trauma processing in the infant brain and this infant system is reactivated in older infants if the mother is present.

I don’t want to get too detailed but I have to say that we know that this process of fear suppression is due to the caregiver suppressing corticosterone levels using social buffering. The data on humans in a clinical situation suggests that social buffering, which involves reducing the stress hormone, is critically important in producing recovery from trauma.

While the infants’ amygdala is not learning to fear the caregiver, the amygdala is responding to the trauma. Here is information about gene expression and a neurotransmitter called amygdala. After just one day experience with trauma, the dopamine and its related gene expression all decreased. This is the normal response of the infant brain. However, after repeated days of exposure to trauma, we have high level of the stress hormone corticosterone. This high level of corticosteron  occurs in abusive relationships, you actually have increasing dopamine and dopamine related genes changes in the amygdala. . This repeated trauma effect in the amygdala is more like the adult amygdala response.  Since dopamine is very important for reward and learning in adults and indicate that future attachment learning will not occur since the amygdala will support fear learning.  Importantly, since the abusive attachment was already learned by the infant, these amygdala changes will not disrupt the attachment learning that has already occurred to the abusive caregiver.  Our data suggests that the there should not be concurrent activation of the fear circuit and the attachment circuit – yet exposure to early life trauma from the caregiver produces this dual activation.  

So it seems bad experiences that occur infrequently do not produce the altered amygdala and attachment learning is preserved. However, if trauma happening repeatedly, it produces an adult-like response in the amygdala.  One additional problem is that this malfunctioning amygdala now communicates to other newly developing brain and influences how these brain areas develop, such as the prefrontal cortex.

And finally, the most pronounced effects of repeated trauma within attachment occur in later life, as has been documented clinically. I don’t have time to show you our data about this but we find later-life depressive-like behavior following abuse attachment.  

One of our most unexpected results relates to how stimuli associated with early life trauma - such as the odor paired with trauma in infancy. In adulthood, when this odor from infant trauma is presented to an animal that is showing depressive-like behavior, the odor normalizes the behavior and amygdala. That is, we can repair this behavior and the amygdala by presenting cues from the early life trauma. So it just illustrates that during the trauma, we might think that the child is not learning but items or cues associated with the trauma within attachment actually become incorporated into the attachment circuitry. So what does this mean? We think that items associated with attachment acquire some global qualities of attachment, such as safety signals, regardless of whether the attachment was associated with abusive or nurturing caregivers.

Here is our take home message. The immature brain processes trauma differently than adults. It is more likely to be associated with the attachment circuit rather than the fear circuit. This occurs because in the very young animals the amygdala fear circuit is not functional. As infants mature and the amygdala can function in fear, the presence of the mother blocks it so these older animals do not learn fear. In other words, the mother has control over the infant amygdala. While in the short-term, this is beneficial to infants because this preserves attachment, there are enduring long-term consequences, including depressive-like behavior. And in adulthood, we find that the stimuli associated with attachment, even abusive attachment, retain powerful control over the brain and behavior - it can rescue some aberrant behaviors and amygdala. These attachment cues appear to work via increasing serotonin and decreasing the stress hormone corticosterone.

As is listed on this slide, tons of people have worked on this in my lab, with lots of collaborators. Thank you so much.