Why our dog’s trauma memories are made to stick

Today, we pick up from the last post which explored why some dogs are more resilient and others are less robust.

Essentially, resilience comes down to three things: temperament, early development and the timing of the traumatic event.

Why is it, though, that my dog remembers the vet’s office even though nothing really terrible happened, and they can’t remember to sit when I ask? What is it that makes them panic when it really wasn’t that dreadful?

For this, we need to have a good grip of some fairly complex neural processes that scientists don’t fully understand yet. Although there are a lot of gaps to be filled in, there’s a lot to help us better understand our dogs.

What learning looks like

It would be nice to think that brains were organised like filing cabinets or digital document folders. In one, we would find learning like sit or walking to heel. In another, we would find the files for waiting before running out of the door.

Looking at the white and grey matter in a brain, it can be hard to think that this mush of tissue is the store of our learning.

Can we identify this or that neuron and know that it contains the memory of our mother’s perfume or our grandmother’s biscuits?

Are the neurons like little filing cabinets, each containing a memory like a video file on our phone?

Truth be told, it’s all a bit more Matrix than that. In fact, your video file on your phone is not a video, really, is it?

It’s a string of code. It’s an electronic representation of moving visual images.

The brain is actually not that different. That is electrical and chemical representations of all kinds of sensory data.

It’s simpler to think of brain cells being more like piano keys than filing cabinets. Canadian psychologist Donald Hebb gave us the helpful line that ‘neurons that fire together wire together’. That’s basically what learning is and what memories are. Like chords on a piano where multiple keys are pressed to create the sound, our memories are probably little more than specific groups of cells firing together.

Wilder Penfield and the burned toast

Since the 1870s, scientists in the West had been cracking open animal skulls and stimulating them with electricity to see what moved. By the early 1900s, we had a pretty good idea that the outer cortex of the brain controlled muscle movements.

Wilder Penfield was an American-Canadian neuroscientist who wasn’t averse to cracking open a skull and rooting around with brain cells. Because of scientists like Penfield, we have an idea about the neurogeography of the brain. Penfield cracked open skulls and systematically applied electrical stimulation to parts of the brain. His conscious patients would then tell him what happened.

As a result of Penfield’s processes, psychologists and neuroscientists were able to create homunculi. These ‘little men’ were visual images of how much of our cortex is dedicated to motor processes and sensory processes.

For instance, the motor homunculus in humans looks like this:

Different parts of the cortex control different parts of the body. We now know that there are two homunculi: the somatosensory homunculus and the motor homunculus. The motor homunculus maps out our bodies; the somatosensory homunculus maps out our senses. They also show us how much space the brain dedicates to each.

Penfield worked on many topics including epilepsy. In seeking to identify the primary location of one woman’s seizures, Penfield used this process of electrical stimulation to identify which sections of her brain were causing the problem. Before seizures, the woman reported smelling burned toast. On national television, Penfield stimulated various parts of her brain until she said she smelled burned toast. He was then able to perform surgery. This way, they could remove the brain tissue where the seizures started and reduce her seizures.

Penfield’s legacy

Hebb, whose phrase ‘neurons that fire together wire together’, was a student of Penfield’s. Their legacy helped us understand that learning is not encoded by filing things away in drawers or folders, but by neurons firing together in specific ways. What we’re learning is that it’s probably the firing of a certain group of cells together that stores our memories.

Learning at a neural level is simply that: neurons firing together.

Memory is simply the retrieval of learning. When that happens, the neurons fire together again.

The more they fire together, the easier it becomes for them to wire together. The more they wire together, the stronger connections they make between each other. They branch out. They make more connections. Like two countries with strong trade links creating a large number of transport routes, the brain does the same thing. And, like geographical trade, the brain does a lot with the bits that are closest to each other.

It also gets easier for those neurons to fire again together. More connections, more growth, more communication. They also sensitise. That means it gets easier for them to fire.

The more frequently those piano keys are pressed together, the more easy it becomes to press them again in the future.

But what presses them? Wouldn’t it be nice if we could develop machines that could stimulate the memories we want to remember and cut out the bits that we do not?

Entering the realm of science fiction

In many ways, one of Penfield’s legacies is to science fiction. In Philip K. Dick’s sci-fi novel Do Androids Dream of Electric Sheep? you may remember the Penfield Mood Organ, a fictional device used to program our own moods, named after none other than Wilder Penfield himself. Wouldn’t it be just lovely to have such a device for dogs to help them feel safe?

Other texts have played with similar ideas. You may have seen Kate Winslet and Jim Carrey in Michel Gondry’s Eternal Sunshine of the Spotless Mind in which characters elect to have painful memories erased. Many of us with reactive dogs long for such devices. Wouldn’t it be nice to erase such memories?

The ability to stimulate the good and erase the bad seems like it would be the ultimate solution to our dogs’ problems.

Since we cannot create a Penfield Mood Organ and we cannot erase painful memories, we are left with one solution. If we know what presses those piano keys together to play the chord, then we can intervene. We can even stop those notes being played together.

In other words, if we know what triggers our dog’s fearful memories, we could really change things. Likewise for your car chasers: if we knew what was pressing those exciting chords, we could really change things too.

Encoding

The brain does not remember all things equally.

If it did, our dogs would be able to recall how to perform pattern games as easily as recalling how to perform barking and lunging.

Wouldn’t it be nice if our dogs could remember to walk nicely alongside us instead of barking and lunging at some dog minding his own business some 500m away?

Some things are more sticky.

Those neurons fire together once and then they’re firing together all the time.

What we also suspect is that learning moves. It gets into the brain at one place and then it migrates.

We know this as a result of human patients like HM. Henry Molaison, HM in scientific studies, had a bilateral temporal lobotomy in 1953. This removed his amygdalae and two thirds of his hippocampi as well as other parts of his brain in order to reduce his epileptic seizures. Henry became the focus of a huge number of studies for many reasons. The reason we’re most interested in was that Henry was unable to pass short-term learning into his long-term memory.

His long-term memories before 1953 were as intact as you’d expect. He could remember some things he’d done as a child, events that had taken place, knowledge he’d learned, experiences he’d had. He could remember things in the short term, like a phone number you gave him. Henry simply couldn’t make new long-term memories after the surgery. There was a lot of damage to his episodic memory for events, stretching back for many years, but the main impact was in his ability to consolidate new memories.

The role of the hippocampus in learning

There are different types of memory that we need to consider when thinking about learning and memory as a whole. You will know yourself from your own life that there are different types of memory. Riding a bike is a common metaphor we use for procedural memory: procedural memories are the memories of how to do things. Even though you may not have ridden a bike for years, the muscles and balance required are one type of memory. The same is true for many procedures or actions, like swimming. Some procedures like baking a cake might need a few reminders after a time gap though.

Declarative memory are the memories we have of facts. This is often what we think of as ‘learning’ even though there are plenty of other forms of learning too. As you’ll know, your recall of facts is the one you think of as decaying most easily, much to the misery of exam candidates across the world.

Henry Molaison helped us understand that the hippocampus is involved in the sorting lots of our episodic learning as it moves from the short term to the long term. Though he recalled how to do things and he could access procedural memories, he could not access declarative memories or emotional ones. Thus, psychologists could understand what experiences were processed by the hippocampus. The hippocampus is heavily involved in two processes.

The first of these is acquisition. This involves acquiring new learning of facts and experiences.

The second is consolidation. This involves laying learning down and storing it. At this point, learning is stabilised and solidified.

Henry’s life also helps us understand that the hippocampus is not involved in Pavlovian processes or in procedural memory, like riding a bike.

The amygdalae

The amygdalae are two almond-shaped areas of the brain that work along with the hippocampus. Although their role is complex, they’re essentially tagging what floats in from the world for processing, particularly deciding what’s a threat to our being. It’s basically like a big alarm button that not only recognises when we’re in a threatening situation that we’ve been in before, but also goes around flagging up new learning for storage under VERY IMPORTANT.

If you’ve ever watched M*A*S*H, you’ll know Radar. Walter O’Reilly, known as Radar, was our amygdalae in human form…. sensing when attack was imminent before anyone else realised and filing everything away for ease of retrieval later. Our amygdalae are there, rubber-stamping our learning to say ‘DO NOT REPEAT!’ and ‘VERY UNPLEASANT!’, making sure we remember that stuff rather than Year 11 Latin.

You might think that it would be quite useful to remove the over-active rubber-stamping advance warning system in reactive or fearful dogs. One thing that has been noticed subsequent to amygdalotomy is that people then become poor at recognising risk, particularly in expressions and body language. Evolution has been particularly effective where the amydala is concerned. Too big and too powerful would make us incredibly risk averse, too frightened to feed ourselves or seek a mate; too small and too weak would win us a place in the Darwin Awards.

The amygdala is mostly involved in consolidating memories, particularly associations. Its strength is in X predicts Y kind of memories, particularly where X predicted a very unpleasant Y. Like Radar learns to identify the distant helicopter sounds as a predictor of incoming wounded, the amygdalae are very good at marking up fearful events for storage. You can read more here.

Why trauma memories are made to stick

When the sympathetic nervous system is aroused (think fight-flight-flirt-freeze) because the hypothalamus woke up properly, our bodies course with cortisol, adrenaline and noradrenaline. Those noradrenaline molecules become office aids in the hippocampus. They are the highlighters, the sticky notes, the rubber stamps. They attach themselves to learning as VERY RELEVANT.

A little stress aids memory. You know this. Whether it’s eu-stress or dis-stress, our hippocampus has noradrenaline receptors that make things more memorable. Noradrenaline is our brain’s sergeant major, rushing around making us pay attention to things and notice other things. It affects our perception and our arousal levels.

As learning is encoded, it’s most likely that the action of the amygdalae and the hypothalamus, specifically that noradrenaline, are sticking to certain memories. They’re amplified and given priority. In the huge orchestral symphony going on in our brains, noradrenaline gives certain chords a VIP ticket. They get a carte blanche to act differently than the other learning we’re doing.

Every time those chords are played, they usually get stronger. They also become sensitive to being played. They get easier to play. In science terms, they’re easier to retrieve.

No wonder, then, that traumatic experiences can be challenging to work with.

Trauma memories in animals

Of course, it is even harder to understand the impact of traumatic experiences on animals, particularly the degree to which memory is playing. We rely on language to understand and intervene with human trauma. What probably happens is that the process of retrieving memories and reliving experiences strengthens them. Every time the world triggers a memory of the experience, it strengthens the neural connections that encode it. This has many implications and it’s a very new science in many ways.

This process, known as reconsolidation, is an exciting process to study. Prior to 2000, we believed remembering played a passive role in learning. We now know that remembering is an active process in the brain. It is a time of plasticity, growth, pruning and shaping. This process is what gives us hope that emotions can change and behaviour can be modified. Since reconsolidation in the mammalian brain is studied across a range of species in the laboratory, we can have more confidence that memories can be modified.

A cautionary tale

At the same time, much of the research happens with rodents, primates and humans. Although these models of study usually help us generalise to other species, we should understand that each species is trapped within their own experiential bubble. As American naturalist Henry Beston said, ‘Animals are not our brethren, they are not underlings; they are other nations, caught with ourselves in the net of life and time.’

The experiences of those other nations can be hard to understand, particularly when there is so little science. We need to be wary of considering our own experiential bubble as the same as that of our companion animal’s. Our dogs may be caught with us in the net of life and time, but their worlds are alien to us in many ways. We cannot imagine, for instance, the ways that dogs process odour, or whether odour triggers memories as it does for us. Certainly, anecdotal evidence would suggest that it might. Likewise, particular sounds.

One dog I lived with came to me from a particularly turbulent home. She arrived anxious, aggressive and on edge along with another dog. When the older dog died, I removed her collar and put it on a bookshelf. One day, months later, I was cleaning the shelf and the tags on the collar clinked. My dog froze, then wet herself.

What did that clink mean to her? Anything at all?

Or am I, in my infinitely human way, associating two disconnected events and finding causality when there is none?

Next time…

In the next post, we’ll look at the processes of acquisition, consolidation and reconsolidation of memory. The process of reconsolidation means memories and experiences open up to modification. While we cannot erase learning, we can modify it. That is something many of us need to hear when we work with dogs who have had a traumatic experience. How we do this and what we do are of huge interest when we work with reactive dogs to change their experiences for the better.