Can Fear be Rewarded?

This forum is all about training and behavior. Everything from potty training to working titles!

Postby maberi » November 13th, 2009, 2:30 pm

This discussion came up on the clicker solutions forum and someone posted the following article. This goes against everything I have ever heard or thought about this topic and I found it extremely interesting and thought provoking.

Just wondering what others think

Can Fear be Rewarded?

Of all the myths and misconceptions that have surrounded dog behavior and training for a very long time, perhaps the most persistent is the one that claims it is possible to reinforce an animal's fear by paying attention to him or trying to reassure him. This was certainly a belief some of Suzanne's first mentors indoctrinated her with close to thirty years ago.

For almost that long, behavioral researchers have known that it is difficult to influence emotional states and involuntary behaviors using rewards and punishments. Ever tried to influence your heart rate or other physiological parameter using biofeedback? Not easy to learn, even given the cognitive abilities of people that animals don't share.

So, if your dog is afraid of say thunder, and you hold your pet and calmly stroke her to help her relax and calm down do you think her fearful behavior is likely to increase? That's what reinforcement does - increase the frequency of behaviors it follows. So if reassurance rewards fearful behavior, then it also follows that your pet is becoming more afraid as a result. It's the emotion of fear that is expressed with observable behaviors. If your pet's feelings of being afraid don't intensify, then neither will her fearful behaviors.

There seems to be a persistent belief that it is possible to reward fearful behaviors without rewarding the emotion of fear. This may be true in people. We all know individuals with a "martyr" mentality who will act fearful and helpless just to get attention and have others take care of them.

This doesn't seem to be true for animals. They don't pretend. If they don't feel afraid, they don't act afraid. When their emotional state changes, so do their behaviors.

Behavioral researchers back in the 1940s, conditioned rats to jump to the other side of their enclosure in order to avoid the shock that followed a buzzer. In the next phase of training, the researchers changed the sequence so that cheese followed the buzzer and the shock was discontinued.

Over multiple experiences with cheese follows buzzer, even as the rats attempted to jump to the other side, do you think the rats became more fearful and increased their jumping behavior? That's what would happen if you believe the jumping behavior (and therefore the fear) was reinforced by the cheese.

Just the opposite occurred. The rats' fear decreased, the jumping stopped and they began to eat the cheese. This is an example of classical conditioning changing behavior by changing emotional state rather than operant conditioning rewarding fear.

Copyright 2007, written by Suzanne Hetts, Ph.D. and Daniel Q. Estep, Ph.D., Certified Applied Animal Behaviorists. Reprinted with permission from “Pet Behavior One Piece at a Time”.
Look beyond what your own eyes see
User avatar
maberi
I Save My Empty Calories For The Bottle
 
Posts: 2781
Location: rochester, ny

Postby mnp13 » November 13th, 2009, 3:52 pm

I don't know what the correct verbiage is then, but I do think fear/anxiety can and is rewarded.

I have seen it a good deal with my "problem kids." For example, the Great Dane in my last class had serious anxiety problems. When approached, she would hid behind her owners, and the owners would soothe and "coddle" her. One of the first things I have owners do at an evaluation is stop talking to their dogs, and to step away when the dog tries to hide behind them. When the "comforting" is removed, the dog acts differently, not a lot differently, but there is a difference. When they make an effort to stop the comforting behavior over the next weeks, the behaviors do seem to decrease.

It seems to me that the dog learns "I'm worried, and mommy gives me hugs and makes me feel better" so every time they feel the slightest bit uneasy they run for comfort. So their threshold for wanting comfort gradually gets lower and lower because they don't have to deal with anything because mommy will protect them from the world... so, yes, I do believe that over time the behavior does get worse.
Michelle

Inside me is a thin woman trying to get out. I usually shut the bitch up with a martini.
User avatar
mnp13
Evil Overlord
 
Posts: 17234
Location: Rochester, NY

Postby TheRedQueen » November 13th, 2009, 4:07 pm

I'm on board with the fear can't be reinforced thoughts.

We discussed this a while back...I'll have to see if I can find the thread. I brought up Patricia McConnell's blog article about reinforcing fear of thunderstorms...
http://www.theotherendoftheleash.com/ca ... a-in-dogs/

I've tried many things with my fearful dogs...and I've found that "coddling" doesn't change anything...doesn't make it worse, doesn't make it better. If anything, it helps me feel better...;) Changing how the dog views the SCARY thing is the best thing I've found for fear behaviors. Inara views the mail-lady as something that makes treats happen now...but I don't know that the emotion behind the barking behavior is changed any

I'm scared of many things...and having people tell me "It's gonna be okay" doesn't make me MORE (or LESS either) scared of something...(though it might make me more annoyed at them). Why would the dogs be any different?
"I don't have any idea if my dogs respect me or not, but they're greedy and I have their stuff." -- Patty Ruzzo

"Dogs don't want to control people. They want to control their own lives." --John Bradshaw
User avatar
TheRedQueen
I thought I lost my Wiener... but then I found him.
 
Posts: 7184
Location: Maryland

Postby amazincc » November 13th, 2009, 4:35 pm

I definitely think fear can be re-enforced/rewarded... Mick is my best example. :neutral:
User avatar
amazincc
Jessica & Mick
 
Posts: 9814
Location: Holding them both in my heart.

Postby maberi » November 13th, 2009, 4:39 pm

mnp13 wrote:I don't know what the correct verbiage is then, but I do think fear/anxiety can and is rewarded.

I have seen it a good deal with my "problem kids." For example, the Great Dane in my last class had serious anxiety problems. When approached, she would hid behind her owners, and the owners would soothe and "coddle" her. One of the first things I have owners do at an evaluation is stop talking to their dogs, and to step away when the dog tries to hide behind them. When the "comforting" is removed, the dog acts differently, not a lot differently, but there is a difference. When they make an effort to stop the comforting behavior over the next weeks, the behaviors do seem to decrease.


Even though I've been in the class and have seen these dogs I will pose these questions because they are what came to mind.

Are the dogs behaving differently because the owners have stopped petting them, or because the dogs are becoming desensitized to the environment and stimuli over time?

If we agree with the article I posted then we will agree that fear cannot be reinforced. So in the example you indicated, the Dane's fear isn't getting reinforced but the running to the owners when fearful is. Is this a problematic behavior? If the dog sees the comforting as reinforcing (which it must if the behavior increases) then wouldn't this therefore reduce the fearful behavior?

I do think that emotions are contagious and that if the owners were feeling anxiety in that situation then they could very easily increase the anxiety in their dog in that instance.
Look beyond what your own eyes see
User avatar
maberi
I Save My Empty Calories For The Bottle
 
Posts: 2781
Location: rochester, ny

Postby maberi » November 13th, 2009, 4:41 pm

amazincc wrote:I definitely think fear can be re-enforced/rewarded... Mick is my best example. :neutral:


How so
Look beyond what your own eyes see
User avatar
maberi
I Save My Empty Calories For The Bottle
 
Posts: 2781
Location: rochester, ny

Postby amazincc » November 13th, 2009, 4:43 pm

maberi wrote:
I do think that emotions are contagious and that if the owners were feeling anxiety in that situation then they could very easily increase the anxiety in their dog in that instance.


I very much agree w/this statement... again, Mick being my prime example... especially at the vet. :|
Once I "got a grip" about the visits he seemed to handle them much better.
User avatar
amazincc
Jessica & Mick
 
Posts: 9814
Location: Holding them both in my heart.

Postby TheRedQueen » November 13th, 2009, 4:43 pm

amazincc wrote:I definitely think fear can be re-enforced/rewarded... Mick is my best example. :neutral:


Why do you believe this? Because he never got better...because his behavior got worse? Most times I see behavior like this escalate because of anxiety on the owner's part. Jenn (Fenella) has come a long way with Murphy (her fearful/anxiety-ridden dog) in class because she's learned to chill out at the other end of the leash.
"I don't have any idea if my dogs respect me or not, but they're greedy and I have their stuff." -- Patty Ruzzo

"Dogs don't want to control people. They want to control their own lives." --John Bradshaw
User avatar
TheRedQueen
I thought I lost my Wiener... but then I found him.
 
Posts: 7184
Location: Maryland

Postby Malli » November 13th, 2009, 4:45 pm

I think it can, and it can't, it depends.
I can only please one person per day. Today is not your day, tomorrow doesn't look good either.
_______________________________________
"You didn't know of the magical powers of the break stick? It's up there with genies and Harry Potter as far as magic levels go." SisMorphine 01/07/07
User avatar
Malli
E-I-E-I-O!
 
Posts: 6341
Location: CANADA EH?

Postby amazincc » November 13th, 2009, 4:51 pm

TheRedQueen wrote:
amazincc wrote:I definitely think fear can be re-enforced/rewarded... Mick is my best example. :neutral:


Why do you believe this? Because he never got better...because his behavior got worse? Most times I see behavior like this escalate because of anxiety on the owner's part. Jenn (Fenella) has come a long way with Murphy (her fearful/anxiety-ridden dog) in class because she's learned to chill out at the other end of the leash.


I believe I "rewarded" his fear by coddling/petting/giving treats/protecting him at the "wrong" time... thus re-enforcing his believe that he was right in fearing people/uncomfortable situations/the vet.
It definitely had to do w/my own fears and anxieties about his behavior (or what "could/might" happen), and he definitely picked up on that.
He did get better once I took a different approach and acted more matter-of-fact... I could clearly see a difference.

Maybe I'm not understanding the article... :|
User avatar
amazincc
Jessica & Mick
 
Posts: 9814
Location: Holding them both in my heart.

Postby mnp13 » November 13th, 2009, 5:22 pm

I don't believe that fear and anxiety are the same thing, one can feed off of the other, but I think they are distinct.

Anxiety can definitely "travel down the leash" no question about it. That's why I take leashes away, or why I tell people to loosen up on the leash, switch a dog to a long line or put the dog on a flat collar. Keeping the owners talking to me, and not letting them talk to the dog or interact the dog helps separate the owner's anxiety from the dog's anxiety.

A LOT of the dogs that I evaluate for my class have owner problems, as soon as the owner takes a breath the dog is fine; a perfect example is the Rottweiler mix I evaluated. Granted, tons of that one was the barrier frustration of the tight leash, the fear involved and some other stuff... but as soon as she was on a long line and not attached to mom (and mom's anxiety) she calmed down and she was happily following me around the ring taking treats, doing simple obedience and letting me touch her head.
Michelle

Inside me is a thin woman trying to get out. I usually shut the bitch up with a martini.
User avatar
mnp13
Evil Overlord
 
Posts: 17234
Location: Rochester, NY

Postby maberi » November 13th, 2009, 6:51 pm

Agreed
Look beyond what your own eyes see
User avatar
maberi
I Save My Empty Calories For The Bottle
 
Posts: 2781
Location: rochester, ny

Postby katiek0417 » November 13th, 2009, 7:02 pm

Dogs know that petting is a form of pos reinforcement. How do they know that? Because petting is usually paired with something else that's good: a sing-song voice, play time, etc. Heck, you've paired food with praise/peeting enough times, so they know praise, petting, etc are other forms of positive reinforcement ...so you use classical conditioning to teach the dog that praise/coddling/petting is a good thing. So, now, you are using what the dog knows as pos. reinforcement to try to calm them down? Really? The dog just knows that what you're doing has been done to them before when they did something you wanted...so, what does the dog do? It continues to do what it thinks you want. It doesn't know you're trying to calm it down...it knows you're giving it something good.

Now, in a fearful dog, say you string the dog up...you view that as positive punishment, right? You're giving something bad, right? That should decrease the fear, right? No, you're not looking at it the right way...by definition stringing a dog up is done to make a dog submissive - to increase a submissive behavior...and submission incorporates a component of fear in it. So, when the dog submits (shows fear) you REMOVE the bad stimulus (stringing the dog up) which should lead to an increase in fear ...That's negative reinforcement...So the dog learns that being a little afraid keeps it from being strung up...it's avoidance...

Now take it a step further, all negative reinforcement starts as escape training (dog shows fear to escape the bad thing)...then turns into avoidance (shows fear to avoid the bad thing)...

I think people are forgetting that within reinforcement and punishment there is positive and negative at work...
"Rumor has it, compulsion is evil."

Katrina
Sacha CGC - Dumb Lab
Nisha CGC, PDC, PSA TC, PSA 1 - Crazy Malinois
Drusilla SLUT- Pet
Nemo - Dual-Purpose Narcotics
Cy TC, PSA 1, PSA 2, 2009 PSA Level 3 National Champion
Axo - Psycho Puppy
Rocky - RIP My Baby Boy
User avatar
katiek0417
pointy ear hoarder
 
Posts: 6280
Location: Glen Burnie, MD

Postby mnp13 » November 13th, 2009, 7:17 pm

So the underlying question is: can fear or anxiety (both emotions) be rewarded or can only behaviors be rewarded?

and how do behaviors that have their base in the fear emotion fit in there?
Michelle

Inside me is a thin woman trying to get out. I usually shut the bitch up with a martini.
User avatar
mnp13
Evil Overlord
 
Posts: 17234
Location: Rochester, NY

Postby maberi » November 13th, 2009, 7:30 pm

Ok, my disclaimer here is that I've had 2 beers and a very stiff bloody mary so if this doesn't make sense I have an out :wink:

Lets say that petting, coddling, whatever you want to call it, does not reduce a dogs fear. Based on Erin's article, cortisol levels do not drop when dogs are pet, but oxytocin and prolactin levels increase (feel good hormones). If this holds true, then coddling or petting really has no effect on the actual fear the dog is feeling, so the petting is irrelavent.

I still do not see how something positive (petting, etc..) could possibly increase an emotion like fear.
If fear is an emotion and reinforcement refers to something that reinforces behavior, how can you possibly increase an emotion (fear) in this case?

Are you saying that a dog is going to increase their fear behaviors to receive reinforcement?


katiek0417 wrote:Dogs know that petting is a form of pos reinforcement. How do they know that? Because petting is usually paired with something else that's good: a sing-song voice, play time, etc. Heck, you've paired food with praise/peeting enough times, so they know praise, petting, etc are other forms of positive reinforcement ...so you use classical conditioning to teach the dog that praise/coddling/petting is a good thing. So, now, you are using what the dog knows as pos. reinforcement to try to calm them down? Really? The dog just knows that what you're doing has been done to them before when they did something you wanted...so, what does the dog do? It continues to do what it thinks you want. It doesn't know you're trying to calm it down...it knows you're giving it something good.

Now, in a fearful dog, say you string the dog up...you view that as positive punishment, right? You're giving something bad, right? That should decrease the fear, right? No, you're not looking at it the right way...by definition stringing a dog up is done to make a dog submissive - to increase a submissive behavior...and submission incorporates a component of fear in it. So, when the dog submits (shows fear) you REMOVE the bad stimulus (stringing the dog up) which should lead to an increase in fear ...That's negative reinforcement...So the dog learns that being a little afraid keeps it from being strung up...it's avoidance...

Now take it a step further, all negative reinforcement starts as escape training (dog shows fear to escape the bad thing)...then turns into avoidance (shows fear to avoid the bad thing)...

I think people are forgetting that within reinforcement and punishment there is positive and negative at work...
Look beyond what your own eyes see
User avatar
maberi
I Save My Empty Calories For The Bottle
 
Posts: 2781
Location: rochester, ny

Postby katiek0417 » November 13th, 2009, 7:38 pm

mnp13 wrote:So the underlying question is: can fear or anxiety (both emotions) be rewarded or can only behaviors be rewarded?

and how do behaviors that have their base in the fear emotion fit in there?


A lot of researchers in the field of emotion won't even refer to anxiety as an emotion - they say it's more of a state of being/physical state....

But let's put it this way, would we agree that phobias are the same as fears?

You have anxiety (which is bad) when you have a phobia towards something (I have arachnophobia, and when I see a spider, I have extreme anxiety)...moving away from the spider gets rid of that anxiety...removal of something bad...so what am I going to do the next time I see a spider? Get away from it...

Now, that being said, let's refer back to the original behaviorists (Watson, Skinner, Thordike), if you will...they referred to behavior as anything observable to the naked eye (that's what set them apart from cognitive psychologists)...so, is fear observable? In almost every case, yes...
"Rumor has it, compulsion is evil."

Katrina
Sacha CGC - Dumb Lab
Nisha CGC, PDC, PSA TC, PSA 1 - Crazy Malinois
Drusilla SLUT- Pet
Nemo - Dual-Purpose Narcotics
Cy TC, PSA 1, PSA 2, 2009 PSA Level 3 National Champion
Axo - Psycho Puppy
Rocky - RIP My Baby Boy
User avatar
katiek0417
pointy ear hoarder
 
Posts: 6280
Location: Glen Burnie, MD

Postby katiek0417 » November 13th, 2009, 7:47 pm

maberi wrote:Ok, my disclaimer here is that I've had 2 beers and a very stiff bloody mary so if this doesn't make sense I have an out :wink:

Lets say that petting, coddling, whatever you want to call it, does not reduce a dogs fear. Based on Erin's article, cortisol levels do not drop when dogs are pet, but oxytocin and prolactin levels increase (feel good hormones). If this holds true, then coddling or petting really has no effect on the actual fear the dog is feeling, so the petting is irrelavent.

I still do not see how something positive (petting, etc..) could possibly increase an emotion like fear.
If fear is an emotion and reinforcement refers to something that reinforces behavior, how can you possibly increase an emotion (fear) in this case?

Are you saying that a dog is going to increase their fear behaviors to receive reinforcement?




Yes, but what Erin's article fails to talk about is that they found the same thing in humans...specifically, what they fail to mention is that too much oxytocin and prolactin can also lead to impotence in men and chronic frigidity in women...doesn't seem so feel good at that point, huh? In addition, while there is an increase in oxytocin after orgasm, if there are negative emotions involved at the same time, then the increase is not significant...
"Rumor has it, compulsion is evil."

Katrina
Sacha CGC - Dumb Lab
Nisha CGC, PDC, PSA TC, PSA 1 - Crazy Malinois
Drusilla SLUT- Pet
Nemo - Dual-Purpose Narcotics
Cy TC, PSA 1, PSA 2, 2009 PSA Level 3 National Champion
Axo - Psycho Puppy
Rocky - RIP My Baby Boy
User avatar
katiek0417
pointy ear hoarder
 
Posts: 6280
Location: Glen Burnie, MD

Postby maberi » November 13th, 2009, 8:23 pm

Fair enough. If we take that all out of the equation, then are you saying that reinforcing a dog that is fearful will lead to an increase in that emotion (fear) or will it increase the behaviors the dog exihibits when fearful (hiding)? If you are increasing the behavior (hiding), then wouldn't that behavior increase in general? Ex. I go home and stand in the kitchen and ignore my dog so they hide behind me to elicit reinforcement (in this case attention)?



katiek0417 wrote:
maberi wrote:Ok, my disclaimer here is that I've had 2 beers and a very stiff bloody mary so if this doesn't make sense I have an out :wink:

Lets say that petting, coddling, whatever you want to call it, does not reduce a dogs fear. Based on Erin's article, cortisol levels do not drop when dogs are pet, but oxytocin and prolactin levels increase (feel good hormones). If this holds true, then coddling or petting really has no effect on the actual fear the dog is feeling, so the petting is irrelavent.

I still do not see how something positive (petting, etc..) could possibly increase an emotion like fear.
If fear is an emotion and reinforcement refers to something that reinforces behavior, how can you possibly increase an emotion (fear) in this case?

Are you saying that a dog is going to increase their fear behaviors to receive reinforcement?




Yes, but what Erin's article fails to talk about is that they found the same thing in humans...specifically, what they fail to mention is that too much oxytocin and prolactin can also lead to impotence in men and chronic frigidity in women...doesn't seem so feel good at that point, huh? In addition, while there is an increase in oxytocin after orgasm, if there are negative emotions involved at the same time, then the increase is not significant...
Look beyond what your own eyes see
User avatar
maberi
I Save My Empty Calories For The Bottle
 
Posts: 2781
Location: rochester, ny

Postby maberi » November 13th, 2009, 8:47 pm

Found a REALLY cool post on this forum but I can't post it!! Curses to the world
Look beyond what your own eyes see
User avatar
maberi
I Save My Empty Calories For The Bottle
 
Posts: 2781
Location: rochester, ny

Postby katiek0417 » November 13th, 2009, 9:01 pm

maberi wrote:Fair enough. If we take that all out of the equation, then are you saying that reinforcing a dog that is fearful will lead to an increase in that emotion (fear) or will it increase the behaviors the dog exihibits when fearful (hiding)? If you are increasing the behavior (hiding), then wouldn't that behavior increase in general? Ex. I go home and stand in the kitchen and ignore my dog so they hide behind me to elicit reinforcement (in this case attention)?





Haven't you ever seen a dog offer up every behavior it knows to get a reward...? Dogs will do that - i've seen people on here post about their dogs doing that. But if you're smart and don't reward those unwanted behaviors to begin with then they won't offer them...

Now, let's talk about your first part...you can use operant conditioning to reinforce emotions as long as you can outwardly observe an emotion (i.e., a behaviorist would say that you can't condition that which you can't see...but b/c anxiety is almost always a representation of fear, a behaviorist would say that you can condition it).

However, I think what people forget is that operant conditioning and classical conditioning DO NOT each occur in a separate vacuum. They often occur together. Take what happens with drug overdoses...

So dog's ARE ABLE to discriminate (and generalize) among situations...
"Rumor has it, compulsion is evil."

Katrina
Sacha CGC - Dumb Lab
Nisha CGC, PDC, PSA TC, PSA 1 - Crazy Malinois
Drusilla SLUT- Pet
Nemo - Dual-Purpose Narcotics
Cy TC, PSA 1, PSA 2, 2009 PSA Level 3 National Champion
Axo - Psycho Puppy
Rocky - RIP My Baby Boy
User avatar
katiek0417
pointy ear hoarder
 
Posts: 6280
Location: Glen Burnie, MD

Next

Return to Training & Behavior

Who is online

Users browsing this forum: No registered users

cron