Book: Teaching with Reinforcement 

Food as a Reinforcer

Author: Kay Laurence

Not all food is equally reinforcing

Food is a Primary Reinforcer: that needs no previous experience to be reinforcing; such as food, security, using instinctive behaviours.
In normal circumstances food, or treats, are probably considered the easiest to use, but the misuse opens many problems.

I feed my dogs raw food and soon discovered that sticking to these principles when delivering a reinforcer became ineffective. Raw meat sticks to the fingers like glue, and if dropped to the floor attracts every atom of dust and fluff in the vicinity. It cannot be stored in a pocket, it is not easy to deliver with speed, and it is difficult to divide into tiny portions. These may be obstacles that can be circumvented, but a good example of my dogs’ main diet not being suitable as a reinforcer in training.

Raw food is an excellent example of a primary reinforcer as 3 week old puppies respond to the smell of warm, raw meat and eat with only a second’s hesitation. The scent awakens a recognition that this is needed, it is not a learned response, it is an instinctive response. In the past when the dog breeders fell in love with specially prepared commercial food you needed to teach a puppy to eat soaked kibble. Go far enough back and relate to the natural process of the pups eating regurgitated raw meat and we have a primary reinforcer.

For the first 7-10 days I feed my pups their warm, wet, raw meat from my fingers. If it is slightly wet it doesn’t stick, and since it has been kept in the fridge I bring it up to “stomach” temperature with a 10 minute soak in warm water. I haven’t quite regressed enough to chew it and spit it out, the simulation of warm, wet and raw works sufficiently well. At the same time the pups are eating from my hands, they are associating human scent with a reinforcer. Licking my fingers is a noticeable reinforcer for my dogs, and they never fail to recognise my scent. My scent becomes a secondary reinforcer.

learned or associated reinforcers (secondary reinforcers); such as a clicker, a “good boy”, retrieving a ball.
Food and the process of eating can carry an enormous amount of baggage. Every now and then a pup or adult dog comes along that does not seem find food a high value reinforcer. Sometimes this can be traced back to illness as a youngster, but very often it is over zealous caring of the puppies where they have been over fed, or fed on demand so that the relief of hunger is not learned through eating. Good quality food cannot be replaced by great quantity. It is normal for pups to be competitive when eating. This begins within minutes of being born where the pup has to fight to find the teats and their first milk. Siblings will jostle and push against each other for a teat which encourages an arousing competition to find a teat and then grip on for dear life. Litters with very low numbers, or even singles may not develop this competitive desire to eat.

From the nutritional aspect the breeder wants to ensure that all pups get an equal chance to eat without the smallest pup losing out. There are some excellent systems that provide each pup with their own bowl replacing the free for all that was the norm. But step back a little further and we realise that bowl feeding is a modern invention, and pups would normally share a carcass, or break off a portion to take away and eat alone. Ideally some degree of competitive and co-operative eating would be normal, and complete inhibition of this may lead to a “picky” eater.

Usually fussy eaters are taught to be fussy. Like many unwanted behaviours a simple single event leads to a lifetime of baggage. On offering the pup the usual dinner, which was refused, more enticing food was substituted and the pup made a connection between the behaviour of refusing and the extra attention or more appetising food that was offered. Pups can refuse, they can have sore gums or have just tried to digest the Sunday Times newspaper or simply be too tired. Never mind, once is not a problem, a series of successive refusals is a symptom of something out of balance.

Pups enjoying a shared bone. They are too young to be able to break off pieces of bone, but can enjoy exercising their new teeth on something natural and tasty with their litter mates.
Stress can also inhibit eating. It is a survival mechanism to keep our resources focussed on fight or flight responses. Full tummies would reduce our ability to do either. Nature withdraws the digestive resources to route to other more important areas. Ever experienced a dry mouth when you are nervous? Try eating a full meal when you are over excited. It is simply the wrong thing to do.

To eat, and enjoy eating, you need to be relaxed and have a sense of security. In fact your digestive process is probably more efficient when in that state. In class we use the natural eating inhibition to measure the mental state of a dog or pup. Pups, or even adult dogs, can be overwhelmed by the class environment and naturally refuse food. This stress level greatly diminishes learning, so trying to train would be ineffective. When offered food is taken the pup is more relaxed, threats have been assessed, arousal reduced. They have a sense of security with their owner’s contact and learning can begin.

If you like to train with food remember that although it is a primary, natural reinforcer a refusal of food or hesitation to take it should be noted. It may be an indication of stress or that the dog has come to expect a higher value reinforcer in connection with that behaviour.

What food?

Small treats are frequently used to reinforce behaviours, especially when building new behaviours. We need certain practical considerations for this task to be effective:

Tasty, easily digested

Wet food is usually more appetising than dry food. Kibble and pre-packaged treats can be easily kept in the pocket for random reinforcement, but frequent feeding over a 10 minute session can leave the dog with a dry mouth and an essential need to drink lots of water over the next hour. Dry food was not designed to be fed in any quantity dry, it is a convenience for packaging and weight. Imagine eating an entire packet of biscuits without a cup of tea?

Raw food has certain practical problems, but small pieces of cooked meats, or specially cooked “cake” (which is often more fudge-like cubes of baked meat, liver, eggs and flour), jerky, sandwich meats are ideal. They are also easily managed in containers and handled without too much stickiness.

The scent of your food will also become a strong cue and motivator for the dog, as will your clothes. As the dog experiences more pleasure from the highly reinforcing training sessions, the smell of your food treats will trigger an excitement, anticipation and increased production of saliva. All contributing to the perfect fore-runner of an interactive session. Your hands will also become focal points from the scent, this may or may not be an advantage, but we certainly use our hands in excess when communicating and this may easily help the dog. Equally your on-board treat bag or pocket will trigger attention. To prevent setting up predictable patterns take extra care to be flexible and vary the routines. If your dog discovers that you only reinforce when you are wearing your training outfit, they can learn that you are ineffective at all other times. Dogs can learn from a single event, especially when the outcome is highly reinforcing. Food is only one reinforcer and we need to have a complete menu of reinforcers available at all times.

Size

This is an important factor when using high rates of reinforcement. We need to be able to reinforce 20 times in one minute, over a period of 10, 1 minute sessions that becomes 200 pieces of food. This high rate is exponentially more reinforcing than one pile of treats at the end of the session. The dogs make a strong connection between the behaviour and the reinforcers, and a high sense of achievement – a primary reinforcer. Commercially prepared treats or biscuits can often be designed for the dog to take away and enjoy at leisure or involve some chewing process to add to the pleasure. These are unsuitable for training purposes unless a time delay is desired. A pup settling down their crate is a perfect situation for a longer reinforcer snack.

Very small dogs and pups will fill quickly for effective training unless their treats are the size of a piece of rice. You can even use the “lick” technique to reduce the portions again. There is cheese in a tube, also paté on your finger. I begin each session measuring a portion for that dog that is never larger than the dog’s skull, i.e; their stomach size. Once that is finished, there is at least a 30 minute rest to empty the stomach before any further training. They may say they want more, but I want an empty tum! You must also reduce their regular meals by an equivalent portion, unless your training sessions are highly energetic and burning up the extra intake.

From practice and experience I find sliced sandwich meats are ideal. Chicken is probably the least offensive, it peels apart easily like flicking coins. Other meats such as liver roll, luncheon meat, sausage types and ham are also good. You can cook and slice your own sausages or frankfurters. Preserved meats often contain a lot of salt to improve the flavour and the dogs can get quite thirsty. We use chicken most of the time and sausages or cheese as reserves. Hard cheese can be chopped up into small half centimetre cubes. Cooked liver and heart is something extra special. You can use the highest desired food at times when focus is questionable. The more fat content the more appetising. One year we had to “recycle” a lamb that had got over fat at the wrong time of year, and the scrap fats were THE Best reinforcers I have ever used, for every dog in class. Such focus! Choosing the right quality reinforcer to match the behaviour in the sufficient amount to strengthen or maintain the behaviour is key skill.

Managing the Food

Store the food in a variety of places, in particular the locations where you will need a quick supply. My dogs like to run around the garden last thing at night checking out the foxes and other night life. It can turn a bit rowdy and coming in to bed can be delayed. On that particular “call in” there are treats waiting in the kitchen, one of the larger, chewy commercial treats, and a reminder of this guarantees no loitering in the garden.

Always have some prepared ready to go from the fridge, and long-life supplies around the house. Learning is not restricted to specific sessions, you will want to be able to reinforce behaviours (reward the dog), especially a pup, to establish lifeskills, many times through the day. All dogs, but especially youngsters, benefit from reinforcement for good behaviour as soon as it occurs.

Pocket mugging can become a problem – especially if you reinforce it. I could ask you why you reinforced it but it was probably one of those single event learning situations that occur when we undergo a single event lapse of focus.

I am guilty of shopping for clothes with suitable food pockets, and have a range of fleece jackets with the pockets chewed out. But I do urge you to check all pockets before washing. Ham flavoured “clean” clothes are not well received in public.

Your reinforcement

This is a tricky area but food preparation and delivery is a highly reinforcing process for some people. It becomes synonymous with caring and demonstrating love for the recipient. At our classes we have professional foodies who regularly exchange treat recipes and thoroughly enjoy the process of finding the very best reinforcers for their dog. At the other extreme are the folk who may remember to buy a tin of hot dogs from the local shop, but forget the can opener!

We also get mixed up between reinforcement and indulgence. I do not think you have to become so scientifically based that you can never, or would never, indulge, but if a dog performs a behaviour to a certain standard and gets a treat, but then repeats the behaviour at a lower standard and still gets the treat we are giving very mixed messages. We have reinforced a variable standard and can only expect variable results. Whilst out on a walk you call your dog and get a tremendously quick response. On arrival the dog gets the treat, the next time you call, the dog hesitates over the very interesting smell, takes the time to cock his leg, then strolls towards you and still gets a treat you have reinforced both behaviours.

“Aaah ….. but he still came when I called him”.

But was it as treat worthy as the first recall? You will still reinforce but this time appropriate to the effort of the behaviour and you will need the self-discipline to lower your reinforcer, and not take the easy route because of that single event loss of focus moment.

Suitability

Food covers a good range of behaviours we may want to reinforce, with the exception of high activity. The dog may need to pant at quite a fast rate and swallowing may be compromised. On occasion when training heelwork with a high head carriage the food can go down to wrong way unless the dog lowers their head.

Food may not be suitable for dogs that become so over aroused at the prospect of a piece of food they are unable to maintain the behaviours. Although the higher the desire for the food the easier it is to generate and reinforce self-control. These dogs would find learning relaxation with food an impossibility.

When it stops being a reinforcer

Food has a limitation. The dog may begin to feel uncomfortable but not be able to stop eating. I notice this especially with clicker training, where the habit of eating after the click is so strong that the dog cannot refuse. The technique of pushing the food at the dog will also cover this state of fullness. Whereas when tossing a piece of food you can observe the dog’s eagerness to follow and eat. Some dogs will always have room for one more treat, but that doesn’t mean it is good for them!

Stress can also inhibit appetite, the dry mouth is a symptom of high levels of stress, and a good indicator that digestion is likely to be compromised as well. A pile of treats sitting heavily in the stomach does not make too good a reinforcer.

If the dog is hesitant to eat, do not persuade or encourage. Take it as communication, and at this time, that piece of food will not be a reinforcer.

Exercise : Treat loading one at a time

To be able to be a skilled reinforcer you will have to learn how to deliver cleanly, with the least amount of fuss or delay.

Whatever treat you are going to use you should be able to have it ready in your hand, one piece at a time.

As soon as you have promised a treat (this can be a click, “yay” or “excellent!”) whilst the dog is in the eager anticipation the treat should be loaded and ready to deliver. 

Practise before you train the dog

Set yourself some exercises to practice this before training begins. You can indulgence yourself with a bowl of peanuts, or M&Ms and practice selecting one at a time, and moving from A to B. (B is usually our mouth).

I am sure you are skilled at loading a shovel full of treats from A and transporting them to B, but we want the process of reinforcing to last a long as possible and that means we deliver single treat one at a time.

If you are not up to eating then use dried beans, or buttons and practice taking one at a time, from A to B. 

Make it close to training

“A” needs to be similar to the training situation.

If your treats are pocket stored, then you will need to be able to get into that pocket quickly, load up one treat and deliver to B.

Don’t expect this to be achieved without practise. If you have to take out a plastic bag and open that as well, then we are not going to be considered first class reinforcers.

For early training days, especially with puppies you need to impress them as to how good you are at delivering those treats.

Consider how easy it is to collect only one piece of popcorn from a bucket …..

0 Comments

Submit a Comment

Your email address will not be published. Required fields are marked *

Get in touch

14 + 6 =

News on courses, articles and stuff you don't want to miss.

 

Woof!