equine clicker training

using precision and positive reinforcement to teach horses and people

ASAT Conference 2016: Dr Joe Layng on emotions

Happy-Animals-Washing

Dr. Layng started his presentation by asking the audience if they thought animals had emotions like they do. Then he asked if everyone thought the person sitting next to them thought had emotions like they did. I think most people raised their hands.  He said these seem like simple questions but they are ones that have plagued science and philosophy for a long time.

It turns out that studying emotions is difficult and that before we can understand them in animals, we need to understand them better in humans. Even using more scientific methods like studying brain activity or measuring physiological factors had not provided clear answers about why and how we feel emotions.

He described an experiment that showed how complex this is. In the experiment, subjects were given a shot of adrenaline that was known to spike in a certain amount of time.  They were then asked to sit in the waiting room until the adrenaline took effect.  At that point, they were asked how they were feeling.  The waiting room was not empty.  In some cases the waiting room was full of people who were tense anxious.  In other cases the waiting room was filled with people who were excited and happy.

What the experimenters found was that the subject’s response to the adrenaline was influenced by the other people in the waiting room.  When asked to describe their emotions, the people in the room with tense and nervous people said they were feeling tense and nervous. The people in the room with the happy and excited people described themselves as tense and happy.  Adrenaline alone was not responsible for their emotions, but it was a combination of adrenaline and the environment.

One of the difficulties is that while we can talk about our emotions and make assumptions about emotions based on observing behavior, there is really no evidence that our “public” behavior is a valid expression of what we are feeling because that information is “private.” This distinction between emotions, which are “private” and emotional behavior which is “public”, is an important one.

While we would like to think that we can learn to read emotional behavior to get information about emotions, even this is questionable. Why?  Because there is no way to accurately test to see if any person’s description of their emotions (the words they use) is the same as another person’s description.  This is because the words we use to describe our emotions are ones that we learned from other people who looked at our external signs and described how we were feeling.

How do small children learn to identify their emotions? The people around them (parents, teachers, friends, etc…) tell them.   “You are crying, why are you sad?”  “You are smiling, are you happy?”  We learn the words for our emotions the same way we learn other words like dog and cat. But while both the teacher and the child can see the dog or cat (they both have access to the same image/object), this is not true with emotions.   You can’t see a “sad” or a “happy” so there is no criteria that can be used to ensure that both the person learning to label their emotions and the one providing the names are talking about the same thing.

The bottom line is that we really have no evidence that anyone else feels anything like we do. At this point he showed a youtube clip of Ludwig Wittgenstein’s beetle in the box analogy (I’ll put the link at the end). Ludwig Wittgenstein was a philosopher who said that we know something is there (emotions), but we know nothing about them.   If we know nothing about them, then we can’t really talk about them, even though we know they are there.

But of course, we do talk about them, and our words come from observed behavior.   But when we talk about them, is it really about how we feel or about what we want?  He used the example of going to the hospital with an injury.  The nurse might ask you for your pain level.   This seems to be a question about how you are feeling (how painful is it?), but it’s really a question about what level of intervention you want.   Do you want something for the pain? Then you are going to give a higher number.

I have to say that when he said this, I thought “oh, I had no idea” and I realized I have always been unsure what to say when asked that question because I have no idea how painful any of the numbers are supposed to be. But it makes much more sense if the question is really about whether or not you need something for the pain.   It’s not about how you feel but about what you want.

B.F. Skinner came to the same conclusion and this is why he said you have to look at external behavior, not try to guess what is going on inside the animal.  This became the behaviorist point of view, which has been criticized for not considering the importance of emotions, but which really just says that we can’t know what is going on inside so have to rely on observable behavior.

If you take this one step further, what determines observable behavior? It is determined by the consequences, which means that in any situation we have to look at the contingencies.  Dr. Layng told a story about going bear hunting and how the excitement about finding and preparing to shoot the bear quickly turns to fear when the hunter realized the gun is not loaded.  Now the hunter is running away from the bear.  The question is often asked if the hunter is running because he is afraid or if he is afraid because he is running.  Dr. Layng’s answer is that he is running because the bear is chasing him!  He’s running because of the consequences.

The idea that emotions are controlled (or at least influenced by) contingencies is supported by some work that was done 1967 by José Manuel Rodríguez Delgado who studied the brain by putting probes into different parts.   He was doing research to see if emotions could be understood by studying changes in the brain so he would stimulate different parts of the brain and see how the animals responded.

In one study he inserted probes in the amygdala of a monkey’s brain and discovered that stimulation in that area would create aggression.   If the monkey was in an enclosure with a rat, it would attack the rat. If it was in the area with another monkey, it would attack the monkey.  But this behavior would change if another higher ranking monkey was also in the enclosure. If this monkey was present, the monkey would go and groom the other monkey when its amygdala was stimulated.  Why the two different responses? It turned out that when the amygdala was stimulated, the monkey would do a behavior that would “make what’s near you go away.”  So if there were only two monkeys, it would attack.  But if there were three monkeys, it would groom, which was likely to make the higher-ranked monkey leave it alone.   The stimulation of the amygdala did not always have the same response. It depended upon the contingency and the consequences.

He showed some examples of CAT (constructional aggression treatment) with dogs where the dogs that had previously barked, lunged, snarled etc… learned to do alternative behaviors when approached. CAT works by changing the consequences so that friendly behavior gives the dogs what they want (distance) and it replaces aggressive behavior.  He also showed an example with an iguana and suggested that one reason CAT works is that it taps into an old pathway (old because it’s shared by dogs and iguanas).  He said “what is ‘inherited’ is not simply species specific typical patterns, but organism environment consequential relation.”

This has been studied by Ann Tierney who says that species typical patterns are canalized operants, that is occasion – behavior -> consequence relations occur almost automatically. Her work indicates that these behaviors have been selected by natural selection and that the best way to understand instinctive behavior is to understand operant behavior. (Canalization refers to the channeling process that happens when one pattern is selected from more variable operant patterns).

This brings us back to brain structure with the idea that brain structures have been selected that support canalization. Or, in other words, we can look at the brain structures and ask about them in terms of consequences.  If you go back to the experiment with the monkey and stimulation of the amygdala, stimulation of the amygdala did not produce the same emotion in every contingency, but it did produce the same desired consequence, which was wanting to increase distance from something.

With this mind, Dr. Layng listed Jaak Panksepp’s 7 emotional systems and indicated what each one was doing functionally. What is the function of ANGER?  – increasing distance.  What is the function of PANIC? – decreasing distance.  Each one can be viewed in terms of what the animal wants, the desired consequences.   This is supported by Bennett and Hacker who stated that you can’t separate brain state or somatic reaction from context (the circumstances under which it occurs).

So where does this leave us? It leaves us with the idea that when we say we feel something, it is more accurately a description of the current contingencies and the desired consequences.   Your emotions are telling you something about your world.   You don’t want to change emotions, you want to change contingencies.

For example, a person’s level of anxiety is determined by the difference between what you need to do and how well you are prepared to do it. If you want to decrease your anxiety, you do something to make that gap smaller. You might prepare more, or change the situation so you are required to do less.

Israel Goldiamond was in a wheelchair in later years and found that the discomfort of sitting in the wheelchair prevented him from working. Or maybe it was that he noticed the pain because he wasn’t working.  How did he know which one it was? He didn’t, but it doesn’t really matter because the solution was for him to set up a different contingency where working was the more likely behavior.   He said that when behavior changes, it’s important to look at the entire picture and see why the existing contingencies have changed and address it at that level.

I think it’s important to recognize the importance of emotions but also acknowledge that we can’t really know what is going on inside an animal or person. Dr. Layng made the point several times that “emotions” are different than “emotional behavior” and that we can show emotional behavior without feeling the underlying emotions.

I remember Kathy Sdao talking about how you could train an animal to assume an aggressive posture without the accompanying emotion. They did this with E.T. the walrus when they put the aggressive “bell” sound he made on cue. Once it was on cue, he would make the sound, but there would be no aggressive emotion behind it.

A few years ago I was at a riding clinic where the instructor was putting a heart rate monitor on horses. There was one horse who had a history of being difficult and when he got “upset” there was no change in heart rate. I realize that heart rate is not the only indicator of stress or internal conflict, but it was interesting that even though he looked upset, his heart rate stayed the same. My mare, who is afraid of noises outside the indoor, showed a clear increase in heart rate when there were noises, which did correlate with her body language.

I think the message here is that we have to be careful about looking at the entire picture before we make assumptions about what an animal is feeling. And instead of thinking about changing the animal’s feelings, it might be better to find a solution to a training problem by looking at the contingencies and see if we can set up a different situation, with different consequences. Perhaps instead of asking “what does the horse feel?” we should be asking “what does the horse want?”

Categories: Uncategorized

Tags: , , ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s