Alyth
Well-Known Member
The pressure motivates, the release teaches....
Positive Reinforcement
- Thread starter GinnyS
- Start date
Booboos
Well-Known Member
OP your objective is to answer your college question.
First understand the question: this is an evaluative question that asks you to compare PR and NR techniques for riding a young horse. The question says nothing about using NR to the exclusion of anything else, but it does seen to assume that NR is the standard approach.
Second define your terms using references from the literature. You cannot compare X and Y unless you know what X and Y are and the lecturer will not know if you understand X and Y unless you clearly define them.
Third chose one of the following possible answers:
1. Deny that NR is the current technique of choice, argue that in fact it's all (or mainly) PR already (this option is, of course a critique of NR and endorsement of PR).
2. Argue that NR is the right choice, PR much less so.
3. Argue that a mix of NR and PR is the way to go.
4. Argue that PR is the right choice, NR much less so.
Fourth, pick one (to start off with, if need be use a second one) practical activity to illustrate the chosen argument above. E.g. standing still to be mounted may be a good one.
Fifth, explain what counts as NR in relation to the chosen activity and what counts as PR.
Sixth, argue your point with reference to the characteristics of the technique, the particulars of the situation and the nature of the animal.
Good luck.
First understand the question: this is an evaluative question that asks you to compare PR and NR techniques for riding a young horse. The question says nothing about using NR to the exclusion of anything else, but it does seen to assume that NR is the standard approach.
Second define your terms using references from the literature. You cannot compare X and Y unless you know what X and Y are and the lecturer will not know if you understand X and Y unless you clearly define them.
Third chose one of the following possible answers:
1. Deny that NR is the current technique of choice, argue that in fact it's all (or mainly) PR already (this option is, of course a critique of NR and endorsement of PR).
2. Argue that NR is the right choice, PR much less so.
3. Argue that a mix of NR and PR is the way to go.
4. Argue that PR is the right choice, NR much less so.
Fourth, pick one (to start off with, if need be use a second one) practical activity to illustrate the chosen argument above. E.g. standing still to be mounted may be a good one.
Fifth, explain what counts as NR in relation to the chosen activity and what counts as PR.
Sixth, argue your point with reference to the characteristics of the technique, the particulars of the situation and the nature of the animal.
Good luck.
fburton
Well-Known Member
I like it - pithy! Parelli?
fburton
Well-Known Member
I'm afraid you may have muddied the waters rather than clarifying. I think you need to be a little clearer yourself about what overshadowing involves - the application of two (or more - but let's keep things simple) stimuli simultaneously. In negative reinforcement, a stimulus is removed when the behaviour we want occurs. In positive reinforcement, a stimulus is added when the behaviour we want occurs. Therefore by definition the two stimuli (aversive and rewarding) can't present at the same time. So overshadowing doesn't apply in the situation of combined negative and positive reinforcement.
Overshadowing happens when two different stimuli - as a rule, aversive - are applied at the same time. The horse responds to the more salient stimulus resulting in a conditioned response to the salient (typically more intense) stimulus and a diminishing of the normal response to the weaker stimulus.
For example: Rider uses hands and legs together. If the horse is ridden in such a way that bit pressure is more salient than leg pressure, hand pressure overshadows leg pressure and the horse becomes less responsive to the leg aids, leading to "dead sides". (This isn't the only cause of dead sides, of course - desensitization is another.)
Overshadowing can also be used beneficially in combination with desensitization to overcome fear of things like clippers. The procedure is described in detail in Andrew McLean's paper "Overshadowing: A Silver Lining to a Dark Cloud in Horse Training". It's well worth reading if you can get hold of it. (pm me if you would like more info.)
I agree with the rest of your post - you make good points.
GinnyS
New User
Dear all,
Again, huge thanks for all those that have responded and offered views on my original question. Of course I have and am researching the subject in the ususal manner but I was just interested in other people's experiences and opinions for my own interest. There are many of you out there with vast experience as demonstrated by the comments provided and I was interested in how many people actually used and understood these terms in general. I am very grateful for all the discussion and it has helped me understand things more clearly and also to be able to discuss the subject more thoroughly both different angles.
Many thanks to all of you.
Again, huge thanks for all those that have responded and offered views on my original question. Of course I have and am researching the subject in the ususal manner but I was just interested in other people's experiences and opinions for my own interest. There are many of you out there with vast experience as demonstrated by the comments provided and I was interested in how many people actually used and understood these terms in general. I am very grateful for all the discussion and it has helped me understand things more clearly and also to be able to discuss the subject more thoroughly both different angles.
Many thanks to all of you.
Alyth
Well-Known Member
I have studied Andrew McLeans articles that are online and I listened to several of his lectures at our last years 'Equidays' here in New Zealand. Much of what he said made a lot of sense. But this (overshadowing) is one of his ideas that I find hard to accept.....imo much of what he says is 'theoretical' rather than 'practical'. I know he has achieved fame as an event rider etc etc etc and I know many riders who I also respect find him inspiring. But I remain skeptical!! I believe in the kiss principle and much of what he says is too ....elaborate? Complicated? Technical? I was not impressed by what seemed to me to be ignoring the horses point of view? A very close hold on the l!!eadrope to me seemed to be claustrophobic for the horse...rather than relaxing him in a tense environment....little things like that made me wonder!!! Then again other people seem to get amazing results!! So who am I to criticize??!! Perhaps his methods are just not for me!!
fburton
Well-Known Member
Hi Alyth,
In the talks you attended, did Andrew not show video examples of the different methods that he uses? I am quite sure he puts the theory into practice on a regular basis at his behaviour centre.
As for overshadowing specifically, it is a very old idea - certainly not his alone - which has been shown to operate in a number of species, including humans, and a wide variety of situations. So why not for horses too? I have myself seen plenty of examples of behaviour in horses for which overshadowing would be a good, simple explanation - and you will surely have noticed yourself that horses tend to 'heed' one 'message' when there are several things going on at the same time.
I have a lot of sympathy for that point of view. What goes through my head while I am actually working with horses is certainly not a detailed, technical analysis couched in obscure psychology terms! I reckon the left half of my brain takes a rest because words seem to disappear when I'm 'in the zone' with horses. However, I have a habit of replaying - like a video - what I previously did with a horse, in an attempt to understand and improve (or just revel in something that went right), and then the technical terms - or at least the ideas they represent - may come into the picture. Doubtless the fact I am a scientist and naturally analytical facilitates this mode of thought. I can appreciate it's not everyone's cup of tea though - and that other people have neither the desire nor the need to do this.
What you say strikes a chord. Watching Andrew McLean work, I get the same impression. He often looks rather 'mechanical' and lacking in 'feel' as he carries out his behaviour modification plans. I do not get that "Wow, what a connection!" feeling that I do from some other practitioners, like Mark Rashid. On the other hand, I do think that he is effective and that horses benefit from his handling - so he must be doing something right, whatever the underlying theory might be! I hesistate to mention this knowing your support for Parelli, but I have seen similar 'mechanical' application of learned principles by Parelli students (and even the Parellis themselves on occasion
) who seemed markedly lacking in 'feel', so it's not just 'academic' horsemen who are guilty of this.
fburton
Well-Known Member
Since Andrew McLean's online articles have come up in this discussion, it might be timely to mention that the one titled "The Biological Basis of Submission", in which he talks about dominance and hierarchy, expresses views that he no longer holds. Andrew now thinks these ideas are not helpful and a distraction from the business of training horses. When I asked him about this article at the 1st workshop of the then-to-be International Society for Equitation Science in Edinburgh, 2004, he said that he would take it off the web if he could, because he no longer believed in the dominance paradigm. In fact, he asked one online magazine to remove the article but they didn't, giving the reason for not doing so as "It is very popular"!
Tonks
Well-Known Member
As you're keen on using McLean's work/text, I will use HIS words to directly critique what you've said above: (I provide a direct quote):
"When positive and negative reinforcement are used CONCURRENTLY, one may overshadow the other, depending on their relative salience. At low levels of negative reinforcement (rein/leg pressure) and high levels of positive reinforcement (jackpotting), positive reinforcement may be more salient, but as pressures increase, positive reinforcement may become less salient. This suggests that positive reinforcement is best used when negative reinforcement pressure have been converted to light signals" (Equitation Science:2010,pg.133).
Hummm.......??
Pale Rider
Well-Known Member
I tend to agree that a lot of trainers can be a bit mechanical. As can the students they teach, but, teaching someone to have 'feel' is probably the most difficult task of any trainer, in fact I'd say it was near impossible to teach. Occasionally you come across a student who really is good and appears to have 'feel' and be in touch with horses, but these are few and far between. I think that you can develop 'feel' but you need to know the mechanic's of the job first. It's only when people start to progress through the levels of these training programs and working with horses that they get these penny dropping moments and insights which help them to develop 'their feel'. One thing I do know is that there is only one way to develop feel and thats with working with lots of horses and putting in the hours. You won't find feel in a book, and no one can give it too you, you need to find it yourself.
Dry Rot
Well-Known Member
Pale Rider
Well-Known Member
Alyth
Well-Known Member
Hi Alyth,
In the talks you attended, did Andrew not show video examples of the different methods that he uses? I am quite sure he puts the theory into practice on a regular basis at his behaviour centre.
As for overshadowing specifically, it is a very old idea - certainly not his alone - which has been shown to operate in a number of species, including humans, and a wide variety of situations. So why not for horses too? I have myself seen plenty of examples of behaviour in horses for which overshadowing would be a good, simple explanation - and you will surely have noticed yourself that horses tend to 'heed' one 'message' when there are several things going on at the same time.
I have a lot of sympathy for that point of view. What goes through my head while I am actually working with horses is certainly not a detailed, technical analysis couched in obscure psychology terms! I reckon the left half of my brain takes a rest because words seem to disappear when I'm 'in the zone' with horses. However, I have a habit of replaying - like a video - what I previously did with a horse, in an attempt to understand and improve (or just revel in something that went right), and then the technical terms - or at least the ideas they represent - may come into the picture. Doubtless the fact I am a scientist and naturally analytical facilitates this mode of thought. I can appreciate it's not everyone's cup of tea though - and that other people have neither the desire nor the need to do this.
What you say strikes a chord. Watching Andrew McLean work, I get the same impression. He often looks rather 'mechanical' and lacking in 'feel' as he carries out his behaviour modification plans. I do not get that "Wow, what a connection!" feeling that I do from some other practitioners, like Mark Rashid. On the other hand, I do think that he is effective and that horses benefit from his handling - so he must be doing something right, whatever the underlying theory might be! I hesistate to mention this knowing your support for Parelli, but I have seen similar 'mechanical' application of learned principles by Parelli students (and even the Parellis themselves on occasion
I'm not very adept at laying out quotes and answers!! But I must clarify the last bit. I do support the ORIGINAL Parelli programme. It made sense and my ponies seem to understand it as well...!!! But I do NOT agree with everything the Ps have done!! And I have heard rumours of other things as well as Catwalk and the Linda levels!!
As to Andrew McLean. At the demos I saw he didn't have video facilities. It was outside and he had a horse and his daughter? handling it.....she held the horse who was quite nervous and upset very close under his chin and that meant it took a while for him (the horse!) to calm down. I would have used other techniques to settle him!!! Andrew demonstrated the overshadowing techniques and I didn't actually agree with him. I felt he was confusing the horse by combining 2 pressures.....personally I prefer to deal with 1 pressure at a time, and get the horse confident before moving on to the 2nd....
As I have moved on from PNH I am exploring a number of different avenues!! I find the basic original PNH techniques are the foundation of what I teach my youngsters....but I want to improve my riding - techniques and abilities (to give you an idea of my level, I achieved Pony Club A level back in 1960 but have lost confidence and I am now .....shall we say.....watching my body condition??!!!...) I have found Heather Moffats Enlightened Equitation the best model so far, but will have to experiment and try out the ideas gradually!! Perhaps I should just forget about improving myself and just enjoy my ponies and riding for as long as I can!!!!
I hope this clarifies where my ideas are coming from!
Tonks
Well-Known Member
I would like to point out that McLean's work uses Equitation Science principles. These principles are not 'methods' or a 'way' of training, like 'natural horsemanship' methods or Classical horsemanship. Equitation science provides a way of gathering scientific 'facts' regarding equine behaviour. Hence, the OP thread which discusses operant/instrumental conditioning. These 'facts' regarding animal and human behaviour provide explainations [and theories] regarding motivation and learning.
I'm sure however, that as we're all human - McLean can only but interpret!
Last edited:
Alyth
Well-Known Member
Yes Equitation Science tells us how horses learn, but then that has to be translated into how we manage our equines....doesn't it????
fburton
Well-Known Member
Hummm indeed. The quote is a caption to what looks like one of Paul McGreevy's figures. It is unclear to me what point the authors were trying to make with it beyond the fact that, at a certain level, aversive stimuli become more salient than positive stimuli. Concrete examples aren't given of how relative salience might make a difference to the outcome in the case of combined negative and positive reinforcement, where the aversive stimulus is not given concurrently with the rewarding stimulus. Can you suggest an example? Unfortunately, there is no context given in the main text beyond "The optimal use of positive reinforcement is described in Figure 8.3." and no examples are offered either, so the figure stands alone and somewhat tantalizing.
(The graph has an odd feature - the "Positive reinforcement jackpots" green dashed oval/zone, extending from low values of positive reinforcement to high, but not to the very highest values. What do you think that means?? I am genuinely puzzled.)
I think the "poisoned cue" phenomenon is important when -R and +R are used together but it seems to me this is quite different from overshadowing. In any case, poisoned cues aren't mentioned in this context, so it isn't obvious that's what the authors had in mind.
The problem with allowing "overshadowing" to refer also to stimuli that don't occur at the same time, as in combined +R and -R, is that this is inconsistent with other explanations - such as why a secondary reinforcer (such as voice) works well when given before a primary reinforcer (food) but can lose its value when given at the same time due to overshadowing. It would mean that, when using combined but non-concurrent stimuli, you could get a salience overshadowing effect in one case but not another. So what other principle would then allow us to predict what will happen in each case? I hope you can see that broadening the definition to include non-concurrent stimulus is liable to be troublesome!
Anyway, I will ask Paul for clarification after Christmas. (He suggested a figure with a strikingly similar form in the paper I coauthored with him a few years ago.)
Tonks
Well-Known Member
Arh.....Xmas over......5Ibs heavier......about £500 pounds lighter of wallet!!
I do agree that the graph and comments do appear to stand alone without clarification and I do agree that examples would be most useful. I can't decide whether this is simply because the author thought it not necessary to discuss concurrent use of PR and NR to modify behaviour, didn't understand it himself (I doubt this) or whether it is simply too complex to get into on this page!!
I must admit however, I am a little confused why, at this juncture, you discuss salience where NR and PR is not given concurrently as I thought the point in question concerned the occurance of NR and PR simultaneously. Although, I do understand that it is the term 'overshadowing' that is the main concern.
McLean DOES discuss [on page 83] the use of NR along side PR and also their concurrent use. He does admit that there is only a limited number of studies that have investigated this but examples of these studies are briefly described to include Warren-Smith's & McGreevy's investigation into the 'blend' of both PR & NR in shaping the halt response.
Don't think they mention poisoned cues at all (interestingly) and certainly not in this context.
I am a little confused and you will have to explain to me why 'overshadowing' [in this conversation] should refer to stimuli that DONT occur at the same time? I do understand that scientific defintions can't be changed or simply made up.
I know that the term 'overshadowing' is problematic in this context as it is normally associated with classical conditioning when compound stimuli present themselves with the conditioned stimulus. Understanding how 'overshadowing' works within a framework of operatant conditioning is more alined with 'blocking'. So, again I do agree that there is little research in 'traditional' psychology which lends itself to McLean's discussion on overshadowing with simultaneous use of NR and PR.
Most research which discusses 'overshadowing' within the context of reinforcement and combined NR and PR is mainly located within equitation science, animal training and comparative psychology I must admit. This does therefore, make me [like you] question McLean use of the term 'overshadowing' with the combined use of NR and PR. How is this supported with research?
However, I do concur with McLean [in my humble opinion] in that overshadowing can and must occur in combined RP and PR with the presentation of singular or compound stimuli, simply because innate behaviour, motivation, learning and real life present situations that are difficult to measure and reseach into classical condition and operant condition are ongoing. Indeed, even understanding how PR and NR works is very complex. However, maybe you will inform me [gently] that I am wrong??
Best wishes,
Last edited:
fburton
Well-Known Member
Apologies for taking so long to get back to you. Hope you had a good Christmas!
Yes, one can use NR and PR together (concurrently), and that's what this discussion has been about. However, the stimuli used for these are not simultaneous. In the case of NR, an aversive stimulus is applied and then removed when the desired behaviour occurs. In the case of PR, nothing is given until the desired behaviour occurs; only when (or after) it occurs is the rewarding stimulus given. So the actual stimuli - aversive and rewarding - would occur at different times (non-simultaneously) when NR and PR are being used concurrently. Therefore, I would argue - but you and others may disagree - there is no possibility of overshadowing in this situation because overshadowing requires the stimuli to occur at exactly the same time (or at least overlap, though note that there should be no overlap if NR and PR used together are performed according to definition).
If the aversive stimulus were to be given when the desired behaviour occurred, it wouldn't be NR - it would be positive punishment. And yes, overshadowing would come into the picture if a rewarding stimulus was presented at the same time as an aversive, the net effect depending on the relative salience of the two stimuli - but we were discussing NR+PR not PP+PR.
Does this make sense at all?
I have no doubt that close quantitative study will reveal additional subtleties (if it hasn't already), but I don't think these can invalidate the general points about operant conditioning made in earlier posts.
Best wishes for the New Year.
As you said, the author(s) do mention concurrent use of NR and PR earlier in the book. Maybe whoever came up with the diagram intended the point to be expanded but this got left out inadvertently. Hopefully, PMcG will be able to clarify.
I suspect most (all?) of the misunderstanding comes down to my poor communication of what constitutes 'concurrent' and 'simultaneous' as applied to NR and PR and the stimuli used to achieve these.
Yes, one can use NR and PR together (concurrently), and that's what this discussion has been about. However, the stimuli used for these are not simultaneous. In the case of NR, an aversive stimulus is applied and then removed when the desired behaviour occurs. In the case of PR, nothing is given until the desired behaviour occurs; only when (or after) it occurs is the rewarding stimulus given. So the actual stimuli - aversive and rewarding - would occur at different times (non-simultaneously) when NR and PR are being used concurrently. Therefore, I would argue - but you and others may disagree - there is no possibility of overshadowing in this situation because overshadowing requires the stimuli to occur at exactly the same time (or at least overlap, though note that there should be no overlap if NR and PR used together are performed according to definition).
If the aversive stimulus were to be given when the desired behaviour occurred, it wouldn't be NR - it would be positive punishment. And yes, overshadowing would come into the picture if a rewarding stimulus was presented at the same time as an aversive, the net effect depending on the relative salience of the two stimuli - but we were discussing NR+PR not PP+PR.
Does this make sense at all?
I have no doubt that close quantitative study will reveal additional subtleties (if it hasn't already), but I don't think these can invalidate the general points about operant conditioning made in earlier posts.
Best wishes for the New Year.
guesstimation
Well-Known Member
Not read the replies but if you are looking into this I highly recommend looking up Ben Hart http://www.hartshorsemanship.com
I trained my pony from unhandled to ridden using clicker and PR in general, over 7 years, Ben is one of the easiest to understand and non-egotystical (made up word but you know what i mean)! person I've come across. Him and Sarah Weston
I trained my pony from unhandled to ridden using clicker and PR in general, over 7 years, Ben is one of the easiest to understand and non-egotystical (made up word but you know what i mean)! person I've come across. Him and Sarah Weston
fburton
Well-Known Member
I met Ben when he was a speaker at an Equine Behaviour Forum symposium in 2001 (I think) and he was a most helpful, down-to-earth fellow, as well as giving a superb talk. Sarah Weston's book "No Fear, No Force: A Guide to Handling and Training Semi-feral Foals" is brilliant imo.
guesstimation
Well-Known Member
It is a brilliant book, my pony's story is in it . So many people have egos and also their way is the only way and don't adjust both Ben and Sarah don't have that about them and it makes a world of difference to the horses they work with and the people that look them up.
. So many people have egos and also their way is the only way and don't adjust both Ben and Sarah don't have that about them and it makes a world of difference to the horses they work with and the people that look them up.guesstimation
Well-Known Member
I keep reading about Andrew McLean but not yet got as far as his books I'm worried they'll be a bit to textbooky for me! If you could recommend one to start with which would it be?
fburton
Well-Known Member
"Equitation Science", co-authored with PaulMcGreevy, is definitely a text book. McLean's "The Truth About Horses: A Guide to Understanding and Training Your Horse" was quite good, in my opinion. His latest "Academic Horse Training Book" focuses on the practical and has had good reviews. I haven't read that one yet myself as I baulked at the price. (It is accompanied by an equally eye-poppingly expensive DVD.)
http://www.equitationscience.co.uk/Shop.html
Last edited:
guesstimation
Well-Known Member
Thank you, yes I've baulked at the prices another reason not got as far as buying one and ensuring get the right one!
Dry Rot
Well-Known Member
Do not neglect your friendly local library who should be able to get you most books through Interlibrary Loan for cost of postage, but you will probably have to wait!
Don't use it and you will most certainly lose it.
guesstimation
Well-Known Member
Thanks DR hadn't occurred to me though I do use my local library I'm a very slow reader so I tend to hang on to them too long!
I'm a very slow reader so I tend to hang on to them too long!guesstimation
Well-Known Member
Just to add you're a star DR just searched local library they can get it will least give me a start and know if I really want to buy it,
Pale Rider
Well-Known Member
I've just had a good read of the equitationscience link. I found it quite disappointing.
Alyth
Well-Known Member
LOL I downloaded and read his articles, got his books from the library and reading it much of it made sense....then I watched several of his demos at our 'Equidays' here in New Zealand....several of our much respected riders learn from him and his wife and I expected great things....I was disappointed! The 'learning theory' may be true and accurate but it didn't translate into good horsemanship in my opinion!! And of course my opinion needs to be respected!!! But his demos didn't seem to consider the horses point of view or give him (the horse) a chance to respond.....they were held close up to the chin....not given a chance to work things out for themselves, or even rewarded when they tried....perhaps it's just because his methods are different to mine....but I couldn't understand his reasoning in several instances....for example the 'overshadowing' method seemed to me to confuse things for the horse.....instead of having to learn and adjust to one thing, they had to cope with 2 things at once!! I really wanted to like and appreciate his methods as I respected several of his pupils, but I'm afraid my opinion differed from his more times than I cared to count!!
Dry Rot
Well-Known Member
Just to add you're a star DR just searched local library they can get it will least give me a start and know if I really want to buy it,
...and the author gets a few pennies through Public Lending Rights which is paid for by late return fines so it is win-win all round....
Some years I even get enough to buy a cheap bottle of plonk from Tescoes!
(BTW, I am not one of the verbose writers on horse training mentioned on here who need to be reported to the Campaign for Plain English
).