Well, I do end up using negative punishment but not by choice. I know, sounds odd. But whenever Lilly does something 'bad' she knows she didn't do it well, and doesn't expect a treat. Like, if I ask to lift her hoof, and it takes her a while to agree. I do click and treat her, but only once I'm holding the hoof. When she knows she did it good; I ask for her hoof, she instantly gives it, or has it already ready for me when I go to the other side. She seems happy,like she knows she did good, and readily takes the treat from my hand. If it takes her a while to lift the hoof and I have to ask more than once, I still click once I'm holding it, but I put it down again after I pick it out, and she just stands there with her head down and won't take a treat from me, I have to literally shove it into her mouth. When she does it well, she is still respectful when taking the treat, but she takes it immediately and gobbles it down. When she does it badly, she won't take a treat even if I offer it, so I do end up using negative punishment that way, in that, I don't give her a treat even though she did lift her hoof. Does that make sense? I can never really remember the difference between them all so sorry if that isn't actually negative punishment.
I kinda train my own special way, and it works for me. I never really know which one of the four training methods I'm using lol!

 

Well, I do end up using negative punishment but not by choice. I know, sounds odd. But whenever Lilly does something 'bad' she knows she didn't do it well, and doesn't expect a treat. Like, if I ask to lift her hoof, and it takes her a while to agree. I do click and treat her, but only once I'm holding the hoof. When she knows she did it good; I ask for her hoof, she instantly gives it, or has it already ready for me when I go to the other side. She seems happy,like she knows she did good, and readily takes the treat from my hand. If it takes her a while to lift the hoof and I have to ask more than once, I still click once I'm holding it, but I put it down again after I pick it out, and she just stands there with her head down and won't take a treat from me, I have to literally shove it into her mouth. When she does it well, she is still respectful when taking the treat, but she takes it immediately and gobbles it down. When she does it badly, she won't take a treat even if I offer it, so I do end up using negative punishment that way, in that, I don't give her a treat even though she did lift her hoof. Does that make sense? I can never really remember the difference between them all so sorry if that isn't actually negative punishment.
I kinda train my own special way, and it works for me. I never really know which one of the four training methods I'm using lol!

That's interesting. That does sound like it is effective.

@rambo99, I think what you were doing was actually positive punishment, meaning you added a punishment instead of taking away something positive. I haven't been real clear on all the differences either, but was reading an article and thinking about the idea of withholding something to teach a horse, and it seemed like something I didn't use.

 

I used negative punishment here the other night when feeding. I use same routine every single time I feed. I have both boys grain in buckets and a wheelbarrow of hay.

I give ice his grain first so he's busy eating ,while I go feed cinder his grain and dump his hay. Went in gate to give ice his grain, he wouldn't let me get to his pan kept blocking me. Well I hauled off and whacked him in the side of his face told him back off.

He got the memo and went about 6 feet back an stood looking at me. I dumped his feed an just stood there. He stayed where he was looking at me when I walked away, he very slowly came up to his feed pan.

After that I went about my business like nothing ever happened. Tonight when I went to go give ice his grain ,he backed off and waited till I walked away. He's young he gets pushy tends to test to see if he can get away with stuff.

It's been really cold he's hungry an in his mind. I'm not getting that feed in pan fast enough. He has all the hay he can eat free choice 24/7.

For most part I use positive reinforcement and use treats. Didn't feel that positive reinforcement was going to work,in the situation I described above. Ice isn't a horse I can haul off an whack and use negative punishment all the time. But in certain instances I do. It gets my point across and he knows I mean it.

 

Yeah, I've always thought of -P as something that's not really useful for most animals(thinking 'time outs' & the likes), but I will frequently use withholding/removing feed - eg. if a horse is too eager, showing 'bad manners' or some such, they are 'blocked' from their feed. Don't know why it didn't occur to me before that was -P. So I do indeed use it.

In another thread we were talking about adding an 'uh-uh' cue to signal the likelihood of a punishment if the behaviour is kept up. That works well with +P & can't see why it'd be any less effective with -P, to make that more effective/practical.

Rambo, 'hauling off & whacking' is an eg of POSITIVE punishment, not negative. Remember, in behavioural terms, positive & negative mean + & -. So when talking punishment, +P is adding something UNdesirable, whereas -P is removing/withholding a Good Thing. So if for eg you just stood there with his feed & didn't allow him to have it until he 'behaved', that would be -P.

 

I thought this was a good example

 

Good comments from @loosie.

I actually think the video mostly shows the use of positive punishment. The trainer pushes the horse away when he comes in rather than just withholds a positive. But maybe three kinds are used together? The positive punishment is followed by a release of pressure, so negative reinforcement. But also withholding the feed with negative punishment.

 

does anyone use negative punishment with their horses, i.e. withhold something pleasant to shape behaviors?

All the time, I think. Per Tom Roberts:

"“It will profit you not” means that the horse will not be encouraged to follow a line of conduct other than what we have in mind for him. We withhold any gain – which means we quietly continue with our demands, whatever they may be. We persist. We quietly persist with our demands. This gentle discouragement of “quiet persistence” is something that horses seem to find irresistible."

I'm not a behaviorist and don't understand how people break things into categories. But...

A horse wants to rush home. Every time he breaks into a trot, you turn him 180 degrees. You don't beat him. You let him choose the speed and you choose the direction. You don't get angry. Don't make him trot faster. Don't "Move his feet!" because the horse is already moving his feet. You just choose to let him use his energy in a different direction. When he slows, you turn toward home. If he accelerates again, you turn away from home. The "reward" is what he wants. You just don't give him that reward.

If Bandit starts bucking, I'll raise his head to reduce the severity of his bucks. But I normally don't punish him. Just wait him out and keep doing what made him unhappy in the first place. "We will not run just because someone else ran", for example. I think of it as, "That is the wrong answer. Please try a different answer." When he picks an answer I like, maybe THEN we'll do a run. Or not, and just do something different.

That may not meet the definition of negative punishment. "This will profit you not", to me, means think about the reward a horse is seeking. If it is being sought in a way you find objectionable, don't allow the horse the reward until his behavior is channeled into an acceptable form. But "negative punishment" may have a different flavor. I don't understand the terms in part because I have no interest in them per se.

 

That may not meet the definition of negative punishment. "This will profit you not", to me, means think about the reward a horse is seeking.

See, that's why I think trying to separate the types is kind of pointless. I have been a little interested in thinking about how they might work.

Yeah, you 2 know me well enough I think, to know I feel otherwise, as to the wanting to examine the 'nuts & bolts'. I just feel that I can do so much better if I really understand & examine the concepts. And as I want to help others do better too, I want to share that. But that's just the way my mind works. I appreciate the 'nuts & bolts' don't matter to all. And some come to the same point in a totally different way. But even still, it basically all comes down to what works for the horse & what doesn't work - or as bsms/Tom Roberts puts it, does it profit him, or profit him not?

And I can well understand the attitude that it doesn't effectively matter whether you're using + or - punishment, or neg. reinforcement for that matter. It's mostly 'academic'. I do however, feel it's important to understand the difference between those 'quadrants' and positive reinforcement, because I believe there is a big effective difference. There are many 'side effects'(as in, undesired attitudes, behaviours, lessons...) that punishment(et al) can cause if not very careful, judicious in it's use. I believe vastly more than the 'pitfalls' of using positive reinforcement badly. So, I believe as well as 'careful' use, I think +R should be the predominant 'quadrant' focussed upon, and the rest minimised or avoided where possible (Not saying you can't create a 'monster' with +R if you don't know what you're doing, consider the 'whole picture'...)

 

See, that's why I think trying to separate the types is kind of pointless. I have been a little interested in thinking about how they might work. But as in @bsms example, it is probably both a positive and negative punishment to keep a horse from heading home.

Apparently I do use negative punishment if I really think about it. I might let a horse graze when I get on, but if he walks off he loses the privilege. So I guess anytime a horses loses a privilege, that would be negative punishment.
I might leave a horse untied while trimming hooves, but if he wanders off I tie him. Apparently all four types or combos of them can be useful.

Personally I don't think any type is wrong if a horse understands it and the overall attitude of the horse remains positive. Some try to say only positive reinforcement is good, but some people also think it's wrong to tell kids they are in trouble and there will be consequences.

 

The four quadrants of behavior modification when understood is a precise and technical description of how we and all other animals learn. The term or concept of behavior modification is important. If behavior is not modified by some event, the event is not part of the four quadrants of behavior modification. The animal's perception of the event is critical to the classification of the quadrant. A perception that is not always clearly known.

Behavior modification is an extremely powerful tool for effective communication with the animal world.

Negative Punishment:

To use an extreme example for clarity, consider locking or taking away the use of a cell phone and/or a computer from a teen aged girl as negative punishment. Examining the strong thoughts and emotions going on in her head should provide a clue to the long term effects it may have on the thoughts directed toward the punisher. The behavior will for certain be modified but in what ways? Will there be new behaviors emerge? There could be pages written on these thoughts. Again, perception is important.

Recent Brain Research:

B.F. Skinner's work was done pre-sixties, long before the recent brain research associated with the four quadrants. Even so, I think it is almost pointless to discuss the application, theory, use of, etc of any training method including of course application of the four quadrants without recognition of the emotional consequences of the training or teaching methods.

This is just so so so important to understand on a deep and personal level.

I do not make any claim to be any kind of an expert in any of this. I have been applying my mind to the topic on a daily basis in the recent months. I studied the topic intently in a psychology class what feels like a century ago.

When I taught my horse to turn his head away from me and more or less forward, I would click and treat the instant he turned away from me even though he could smell the tasty morsels in my treat bucket.

So.........was my withholding of a treat until he turned his head negative punishment for his mugging of me? Some might argue it was. I don't think was. If the horse could talk, his opinion would matter. But the click and treat when his head was in the chosen place was almost certainly a positive reinforcement normally referred to as a reward. He very quickly learned that to get a treat his head needed to be forward. (but not his eyeballs-so funny)

The quadrants can become quite a bit more messy in real life than they are in a book on paper. There has to be a real interest and a fair bit of understanding of them prior to going to the field with them to prevent every thing from falling into disarray.

Edit: I think one of the major problems, and I do think it is a problem, of most applications of all training/teaching methods, including the four quadrants, is a looking toward the outward expressions of following behavior. Yes, the all important behavior modification.

But what is just as important, and maybe more important in the long run, is the resulting modification of the unseen inward modifications of the emotions which has been historically mostly ignored with the horses.

Because of their high alert/fear factor, I'm beginning to think the horse is one of the most emotionally sensitive animals on the planet.

 

I think technically yes, withholding the treat until a behavior happened would be negative punishment as @ACinATX describes.

Something those who believe in only using positive reinforcement seem to believe is that the other types are aversive to the horse so should not be used.

My opinion is that we don't have to be that sensitive if the horse isn't.
Horses are used to tolerating a lot of mild unpleasantnesses as part of life. They taste weeds they don't like. That doesn't put them off eating grass. Their friends kick and bite them. That doesn't stop them from hanging out and mutual grooming.

I think horses develop problems from severe or unjust punishment. They stay friends with us if we do things they mildly dislike at times, if the overall relationship is positive. There is no relationship in life that is completely positive.

 

If Pony wants attention but he's being too pushy and in-my-face about it, I turn away from him. If that doesn't work, I walk away.

I think of it as negative punishment because if he wants attention and is being nice about it, I will give him attention, so I'm withholding it in this case.

I guess a more clear case would be if we're doing some sort of training and he's getting rewards for doing the right thing, and then he does the wrong thing -- no reward. Usually in that case I just stand there for a few seconds, up to half a minute, until we reset and try again.

He definitely understands it and does not get worried about it. But he's not a worrier. I'm sure it works for him because every few training sessions he forgets that "hooooo" means to totally freeze, not stop and then turn to me to get a treat. After a few times of withholding the treat, he clearly remembers. You can see him thinking about it. In this case, he'll stop, start to turn toward me, then you can see the little lightbulb go off above his head, he stops turning, goes back to his original position, and waits. And the next time after that, he will just freeze, not turn at all.

 

Yes, @loosie, good points. It can be very helpful to have an ideal of using positive reinforcement the most. I do.

There is a trend for people to feel the need to take it to extremes, which is too cumbersome to work for more than a fringe group.

It reminds me of a big name trainer explaining how anyone could have a horse that followed the twitch of the finger, because his horse did. Then he mentioned that he had spent hours each day for many months tuning those responses.
A) most of us have other jobs and responsibilities, and relationships other than with one horse. B) is the horse truly better off having to work on training for hours each day rather than enjoying a regular horse life?

I have a hard time finding the desire to turn something like moving away from leg pressure that is so direct and simple, and apparently not very aversive to a horse into something very complicated and time consuming to teach.

I don't need a clicker because horses are smart and understand "good boy" and "ah ah" very easily, and every horse I've known wanted to get the "good boy/girl" as soon as they knew what it meant.

Horses actually have a desire to please us when they like us, so things don't have to be a bridge to a treat, or release. Rather, horses can be very pleased just to get a "well done" from us. They really enjoy teamwork and part of riding is feeling the pleasure the horse gets from accomplishing something with you.

 

Rather, horses can be very pleased just to get a "well done" from us. They really enjoy teamwork and part of riding is feeling the pleasure the horse gets from accomplishing something with you.

Yes, this is something I didn't understand about Pony for a long time. Either it was because I was too green, or maybe I was being inconsistent in saying "good boy," or maybe it was because we didn't have that relationship yet. With Teddy, it was clear that "good boy" was a huge reward for him. I think part of that is his anxiety about getting things wrong, though. But it took a while to realize that Pony liked to be a good boy, too. I think that really changed our relationship -- me realizing that he wasn't in it just for the cookies.

This reminds me of a thread I've been wanting to start for a while, about styles of learning. Maybe today is the day -- it's cold and nasty outside and I could either do that or clean the house. Yikes!

 

In a single interaction you can cycle through multiple forms. The grazing example is a good one really - are they being obnoxious and pulling you around while ignoring you? Prod/tap them on the side until they pay attention. Don't let them graze until they are focused on you. If they ignore you and pull their head shake the rope until they lift it. When they finally focus on you praise them and let them graze when they do - with a treat or a pat. I think positive reinforcement is so good because a lot struggles owners have with various animals and personalities is usually because the animal is either uneducated or insecure in their environment/handlers. Positive reinforcement fixes many things... fills many holes. With an insecure or confused animal positive/negative punishment might open the door initially but often doesn't get people very far, especially as many don't even understand them much less practice them constructively and fairly. Positive punishment in some cases also increases the risk of conflict but sometimes conflict is necessary for change.

For your example gotta is the behaviour due to anxiety? Are they hangry? Are they simply so excited that they are distracted and forget you're there? Has the animal actually been taught any manners in the first place (are we expecting too much?) Do you know the animal well and the extent of how well it knows/trusts you? The answer is different for all of those. For an anxious animal I have done a few things - fed them tiny bowls one after the other. Handfed them from their bowl. Given them half first and then did training with the second half 20mins later. From a safe distance or with a stick refill their bowl handfuls at a time. Sometimes doing nothing is best and over time the behaviour will subside as they become more secure. Also consider that if your relationship is thoroughly developed in other areas that the behaviour might naturally correct itself without needing to focus on it.

For a confident animal I know well and taught, I will use positive punishment (a tap or shout "hey!") as an initial warning because they aren't dumb and know better. I would be VERY cautious using positive punishment on any animal I've not developed that relationship/boundaries with or who doesn't know what is expected of it in that given situation. Positive punishment in my personal opinion can have faster and better results when there is a trusting relationship which might seem contradictory at first. An educated animal, ordinarily polite, that just forgot themselves? A smack on the rump, a pointy finger, a loud "NO!" is sometimes all that is needed, followed up by praise when they show you they remembered what IS expected of them. They have relationship with you and know that this is merely your way of expressing displeasure to XYZ behaviour. They can learn because they are not frightened and I consider it, in this context, like a reverse clicker if that makes sense :P Positive punishment and fear don't always go hand in hand - but often does, in the wrong ones. Sometimes the animal is so sensitive that just a look or growl is enough. I find the error is in how much energy or the form of positive punishment, not the intent itself. "I want behaviour to stop" is what it is, at its core. Negative punishment is usually safer for all parties but that is not to say it is kinder, either. It's also important to try have a set sequence when it comes to all this. Don't lead with positive punishment into reward/negative and suddenly next time lead with negative into positive. Spaghetti training leads to spaghetti results!

One is not worse than the other. But some are easier to wield, so to speak, for people that are lacking in either knowledge, eye, feeling, so on so on... positive is simpler, safer and kinder. Think about ALL the people out there that only try positive reinforcement and fail. It's not always because they are doing it wrong. It can be because the "regime" doesn't meet the needs of the individual. Its like taking lots of calcium and paying no attention to D3.

 

Think about ALL the people out there that only try positive reinforcement and fail. It's not always because they are doing it wrong. It can be because the "regime" doesn't meet the needs of the individual.

Great post but I'm confused about this part. My belief is that if +R is properly applied, given the situation and the horse, it just works. Particularly not understanding ( "regime" doesn't meet the needs). Could you help me on this?

 

Right, @Kalraii. Like with an unhandled mustang you do not want to use positive punishment because all they have is fear and no trust yet.

 

And yet, soooo many cop 'round penning' as their first real interaction with humans - no wonder horses often don't want anything to do with people...

 

When a child or friend is doing something they know you don't like and you say, "Oh stop it!", I do not classify that as positive punishment even though it may technically modify the behavior. RE: It gets messy in the real would.

 

I like the "This will profit you. This will profit you not." approach combined with "Quiet persistence."

First, it eliminates extremely common punishments like, "Work him until he can barely breathe!" - used on our pony Cowboy by someone with decades of experience. Or "Hit him with a crop! Show him who the boss is!"

Second, it eliminates the idea of "bad behavior". There is only behavior we want to see more of and behavior we want to see less of. Training for or against. No morality is assigned. No "bad horses".

Third, it makes us think about what the horse wants. WHY does he do something? What is his internal "reward"? If you try to impose a new behavior without eliminating WHY the behavior first started (or was reinforced), you are giving contradictory signals to the horse - the training equivalent of pulling on the reins and kicking the horse at the same time.

And best of all, that simple concept seems to work very well. Every time I've tried it - IF I can figure out what the horse is getting as his own, internal "reward". It forces me to look at every issue from the perspective of the horse - if I can. That is the hardest part. But I'm sooooo tired of people who won't ride without a crop, or who act as if every issue can be solved with a whip. Far too many of those folks in Arizona, including many who have spent decades "training" horses.

Combined with "Give & Take" - do something I want and then we'll do something YOU want - it seems very helpful. Seems to work well with kids, too. Wish I had thought of things that way when I was first a parent....

 

I like the "This will profit you. This will profit you not." approach combined with "Quiet persistence."
...

There are many posts here that I sorely wish I could do more than 'like', and this is one of them - very well & simply put.

 

No, Negative Punishment is too abstract a concept for animals to understand. Hell, teenagers don't seem to get it, how could a horse/dog/parrot?

 

See above. It worked on my Pony. Like I said, I could see him thinking it through and then changing his actions. He's a smart guy, though. I think the issue is that they have to already understand what it is you want them to do, and that they will get a reward for doing it. If he more or less understands that "hoooo" means to freeze in place, but he just tends to forget, then he can be reminded with negative punishment. But if he had no idea what "hoooo" meant, let alone that he would get a reward for doing it correctly, then of course it would be useless.

 

....................

 

I'm curious, does anyone use negative punishment with their horses, i.e. withhold something pleasant to shape behaviors?

As I understand it, there is positive reinforcement, negative reinforcement, positive punishment, and negative punishment.
I definitely use the first three. Positive reinforcement, that is something I use a lot: treats, an encouraging voice, a break to eat grass. Positive punishment I use too: the horse threatens to kick, I give a slap and use a negative voice. Negative reinforcement or pressure/release is also something I use a lot: the horse won't walk forward, I apply pressure on the halter, when he walks forward I release it.

However, I don't think I ever use negative punishment. Do other people find it actually useful? What has kept me from using it has been my thoughts that the horse might not understand it, and that it might not be fair.

For example, a horse that pushes into the handler on the way to their food. I guess some people might withhold the food until the horse can do it nicely. But I would worry that might increase the horse's anxiety and make him even more likely to worry his food may be withheld. I also don't want the horse to think I am competing for his food with him.
What if the horse doesn't behave nicely, then will I not feed him? What if this causes extra acid, gut pain and ulcers?

In my mind it would actually be better to let the horse know you disapprove of the behavior with positive punishment and then give the food right away after he understands he shouldn't push into the handler. Or even better, teach him to stand waiting to one side by teaching him with a halter on and treat reinforcement. Then the final reward of his dinner.
Are there scenarios where negative punishment is a better tool?

Well. Whenever Maverick chews his rope I take it away from him. He chews it less frequently now. But it took a long time for that to work. So I don’t think it’s a very good method, if I’m honest.

 

Reading replies to this thread, I’d like to say that I’ve often also thought about how sometimes an action can be multiple training tools at once. For instance, Maverick used to have a food “aggression” problem (i put quotes because I hate using the word ‘aggression’ because I don’t like using labels like that, but for the sake of not being here all day... y’know). Of course, that wasn’t really his problem. The family always gave him treats and he would come up to them pinning his ears and they would get scared and hand him a treat and back away. I asked them to stop that, that they were training him to be aggressive, but no one listens to the kid.

Anyway, to work on that issue, I brought a bucket of grain and treats into a roundpen, set it down, let him have a few bites, and then asked him to move away from it. He would pin his ears when I would try to do this, and I would increase pressure by becoming “bigger” looking by standing tall, and becoming a bit scary by locking in on him in a predator-like way, and more or less charging him. That could be considered positive punishment for pinning his ears at me, and if he turned to kick at me, he got a smack on the butt, which is also positive punishment. And not having his food anymore would qualify as negative punishment. I would then roundpen him for about 30 sec to a minute, however long it took before he became focused on me and not the food, and once he would do that, I wouldn’t really stop him, just stop driving, and back away, and let him have his food back if he wanted it. That was both negative reinforcement (due to my release of pressure of backing off of driving him) and positive reinforcement (letting him have the food back).

I would like to say, that I don’t like being that aggressive, and so I didn’t keep up with that method for very long even though it was working. Instead, once I felt it was SAFE to do so, I started making him wait before I poured his grain in his bucket. By this point, I had taught him a “stay” command— that was something that I taught him for safety reasons, for both of us. So I would approach with the grain, tell him to stay, and if he started getting antsy, I’d pinch a pressure point on his chest and have him back up, which could be seen as positive punishment, but it was more just to put him back where he started. He may have viewed it as punishment though. When he would straighten up and stop pinning his ears at me about it, I would release that pressure, which would be negative reinforcement, and then when he was still, not earpinning, and not acting impatient, I would pour his food in,make him wait another moment and not rush to the bucket, and then, when he’s still being still, I would then retreat and let him have his food— negative and positive reinforcement at the same time.I don’t feed grain anymore, but it really did help keep him nice and quiet and pleasant when I would feed him.

In my opinion, none of these methods are mutually exclusive nor are they meant to be used that way. They’re each just a tool in the toolbox.

 

It’s an interesting concept and is rather like depriving a teenager of their iPhone because they didn’t do their homework.
Horses don’t have the same thought process as a human though so I’m not sure it’s effective in the same way.
Horses live in the moment, using the food aggressive thing as an example, removing the food isn’t teaching the horse a lesson not to attack you, it’s breaking the cycle of habit.

The scenario of ‘human feeds horse, horse attacks human because he doesn’t want to share food with human’ becomes a learned habit because it’s been successful.

Removing the food at the first sign of aggression also removes the horse’s need to drive the potential food thief away. No food, nothing to steal.

It might work if it breaks the cycle that the horse has gotten into but if the horse still doesn’t respect it’s human and still regards the human as a food thief, the old habit will soon return.

Horse’s have a better understanding of punishment in the way they deal with a potential equine food thief - the horse with the stronger personality will drive the other horse away.

I’ve generally found that food aggressive horses learn much faster if their human asserts their leadership and drives the horse away when it attacks.
You have to mean it though.

 

It’s an interesting concept and is rather like depriving a teenager of their iPhone because they didn’t do their homework.
Horses don’t have the same thought process as a human though so I’m not sure it’s effective in the same way.
Horses live in the moment, using the food aggressive thing as an example, removing the food isn’t teaching the horse a lesson not to attack you, it’s breaking the cycle of habit.

Yes, horses being 'in the moment' beasties, won't likely learn from abstracted 'consequences' - eg. I am losing my phone because I didn't do my homework. That doesn't matter whether we're talking -P or otherwise. But assuming you find an effective -P for the situation/horse, I don't see why they cannot learn, to use the same eg, 'I am not getting my phone because I am not doing my homework'. IOW, it is instant association, and if they DO get their phone when they are DOING their homework, the absence will be strong(especially if somehow phone makes doing homework easier, better).

In the same manner, if every time the horse goes to attack you, you remove their feed, and only allow them the opportunity to eat when they are being 'polite', yes, they will learn not to attack you. I'm not getting why you think that cannot happen.

 

I think negative punishment can seem to work in circumstances where it's so short-lived that it practically becomes positive reward and the line between them gets fuzzy.

I used to do morning grain at a barn where a very pushy gelding would always try to get into his grain before I even got it into his feeder. I would say "in the corner!" and push him over there, and not pour the feed in until he was in the corner and not trying to stick his head in the bucket, then I would pour the grain in, and say "okay!" and step back.

So in that case, is withholding the grain while he is being pushy a negative punishment, or is giving the grain when he does a desired behaviour (stands in the corner) a positive reward? Is it both? Does he understand the negative punishment part, or does only the positive reward part actually compute for the horse? I don't know.

 

So in that case, is withholding the grain while he is being pushy a negative punishment, or is giving the grain when he does a desired behaviour (stands in the corner) a positive reward? Is it both? Does he understand the negative punishment part, or does only the positive reward part actually compute for the horse? I don't know.

To that, I'd say yes to all of the above. It is, as in most situations, not 'clean cut' but a combination. And is an eg, IMO, where it simply doesn't matter whether it is one 'quadrant' or another.

 

+R trainers all recommend using protective contact when working with any animal that has any prosibility of exhibiting dangerous behavior. Simple.

 

This whole discussion (and I admit I haven't read every post, heh, but I'm working on it) is reminding me of a Warwick Schiller video. (I know, I know, everything does!)

One of his principles is "Don't say don't; say do." Meaning that, saying "don't do this thing" to a horse is a whole lot less effective than saying "do this thing instead." As demonstrated here. Instead of saying "don't get into the food while I'm serving it," he has her communicate "DO go stand over there!"

 

Warwick Schiller according to my perception is making a gradual transition to reward based training without scaring all the people off by using the +R jargon.

The +R people would say to reward an incompatible behavior until it has a stronger reward history than the unwanted behavior. By incompatible behavior is standing over there is incompatible with being over here and mugging for food while we feed.

It's hard to teach them NOT to do something without punishment. But teaching an incompatible behavior with reward is easy.

Shawna Karrasch simply teaches all the horses she works with to target and hold on a stationary target with many hanging around in various locations. If she needs to clean a stall or what ever it's just, "Go Target" and they are out of the way for cleaning, feeding, or whatever.

Train to 'go stand over there'. Same thing.

 

I understand looking at "conditioning" as mechanistic. I have felt the same for most of my life. I argued the same point in psychology class.

But I've moved past that objection. I really really believe it is an efficient and powerful way of communicating with the horse in a way the horse actually understands what we are saying in that way.

I would have been just as vociferous or more than you at one time. But the concept of the horse learning they are allowed to say no but wanting and choosing to say yes is too powerful for me to turn away from.

That concept is something I have deeply desired since my first horse. +R to me has shown that it is clearly possible. That desire in me and the possibility of obtaining it is the driving force for the path I have chosen to follow. It is simply impossible to maintain a horse below a critical or evasive threshold using negative reinforcement. I am not saying -R has to be cruel. It has a bad rap from the past but many are now way below cruel with it. But it simply does not represent the relationship I desire having with a companion animal.

 

There are many correct ways. Positive reinforcement training is great. It is definitely not the only method that creates a horse that sees their human partner with positivity and true affection.

What you are inferring is that only those who have used purely positive reinforcement methods have a truly good relationship with a horse. Also that those of us who use mixed methods do not give the horse the choice to say no and choose to say yes instead.
Not to mention, I believe every training method, even +R involves coercion of the horse. A horse will learn that you won't leave them alone until they do something for you, and that is part of why they choose to say yes. It is not that they care more for their humans if they use positive reinforcement versus other methods.

I will submit that no matter what type of relationship you have with a horse, their relationship with their best horse friend will be better. That relationship will not be based only only positive reinforcement, because horses don't operate that way with each other. As I've said, no relationship with another creature can be only positive. It is simply impossible. Even if you believe you will only present positive things to a horse, their feelings can still be negative at times. Even the person at work who always acts super nice can get on your nerves. They might get on your nerves more than the person who is more real and straightforward, who simply says "cut it out" and moves on rather than bugging you all the time.

 

BTW - my horses stay calm at feeding time because they don't like getting hit with buckets or chased around by a near-psychotic bifocal-wearing old man....

 

I'm more with @gottatrot on horses pleasing humans. I think they can find us sources of pleasurable social interaction. Once Bandit and I get out on a trail, he seems to enjoy our working together as a team. I believe horses come to really enjoy a feeling of togetherness with a human and that pleasure becomes a positive reinforcement. But in most cases, a horse will chose the HERD over the HUMAN. And I don't blame them. It is who they are.