På engelska, jag klara inte att lyssna på engelska och anteckna på svenska...Och som vanligt är det mina personliga anteckningar som jag hoppas är förståeliga också för någon som inte är jag.

Ken Ramirez:  Become the best trainer you can be

General aspects of training: species does not matter. Training is the same – we all learn the same way! Of course species are different, and you need to know that, but the training is the same.

 

Goals for the day:

1.Training philosophy of Ken

2.Cross-species training: what can I learn from zoological training

3.Dealing with mistakes in a positive way (mistake an animal make)

4.Advanced tools

 

1.

What is training?

Definition:  Training = teaching! Teach the animal how to live in our world. Teaching is a shared process: we respond to the animals, the animals respond to us. It is a natural process, the mother teaches the children, animals always learn! Our responsibility is to teach them how to have a good life in OUR world.

 

Cornerstones of animal care:

1.Health care (veterinary program)

2.Nutrition (food and vitamins)

3.Environment (includes social structure) Ex: cats like to scratch and climb, get them cat-trees indoors. Horses are made to live together, allow them to do that.

4.Behaviour management (training and enrichmenmt)

All 4 are necessary! Training is JUST AS IMPORTANT as the others.

 

Primary reasons for training:

Anything that benefits the individual.

1.Physical exercise

2.Mental stimulation

3.Cooperative behavior

 

Secondary reasons for training:

Education, Research, Conservation, Entertainment, Work animals, Sport, etc…

As a trainer my first responsibility is towards the INDIVIDUAL! The animal needs must come first! Get your priorities in order.

 

The rules of learning is easy, but behavior is complex.

 

 

 

2.

Cross species training. How hard is it? How critical is it to understand the science? When do differences impact training?

Everything starts with GETTING TO KNOW THE ANIMAL!

 

Species:

• Natural history of species and race – what are they made to do? Herd? Track? Hunt?
• Biology and physiology – how do they naturally act? If you try to teach a dog to fly you will fail, no matter how good your skills as a trainer are.
• Uniqueness and specialization. Dog vs wolf, tiger vs lion, river otter vs sea otter(nosetarget or pawtarget). Similar – NOT THE SAME!

Easier to start training a natural behavior and get in on cue. When the animal understands the training game you can teach complex things.

 

Individual:

• History – acquisition (mother-raised, hand-raised, rehabilitated)
• Previous human interaction (trained before, used to people, bad interactions)
• Observations and ethograms (when they are not watching) – what do they like to do naturally?

 

Knowing the animal

• Diet and nutrition – healthy animal is happy animal. Base diet should be received every day, no matter of training. Depravation is NOT positive reinforcement training. Vary the food so that the animal has a variety of reinforcers you can choose from.
• Create a comfortable environment.
• Develop trust and relationship. Reinforcement history is an aspect of building relationship – we like those who give us stuff.

 

Trust and relationship building

Determine type of contact. Fences and protection? Six types of contact:

1.Free contact

2.Semi protected (muzzle, leash)

3.Protected contact (other side of a fence or other barrier). The animal chooses to stay or leave, you can still touch some part of the animal through a window (ear, snout, paw, etc.)

4.Confined contact (the animal can not leave)

5.Hands-off training (fences or barriers, no physical contact) You can train without touching!

6.Remote training – designed to NOT build relationships, no human involved. Prepare for reintroduction in the wild, or create an interest for the environment. Reinforcement will come from the environment, by remote control or similar.

 

Developing a relationship.

Animals learn 24 hours a day! USE that! PAY ATTENTION!

• Every interaction has a reinforcement value
• Passive vs formal training. Formal: “real” training with clicker and treats. Passive: everything else. When does a puppy start to learn? Right away!
• Compassion (love) vs science. You have to have BOTH!
• Trust takes time!

 

Getting started (trainingprogram)

1.Choosing the right reinforcers

2.Select an event marker – audible, tactile, visual. Not necessary, but helpful!

3.Operant or classical conditioning?

4.Pick the right place to begin. Goal? Starting point for behaviour?

 

Case study 1: targeting

• Choose a target (soft, hard, big, small)
• Look beyond traditional targeting
• A truly target trained animal knows more than basic targeting

Exemples: Nosetouch, targetstick, pawtarget, glove/perch for bird, mat for dog, freezetarget, follow target, grab target, experienced sleddogs is a target for youngsters, stretch out paw through window, laserpointer

8 basic target skills:

1.Basic target (first we touch mark and feed, then we wait for the animal to touch and we mark and feed)

2.Different types of targets (stick, plate, hand etc)

3.Complex (use the target to teach something else, let the animal reach further, search for)

4.Extended (freeze)

5.Follow

6.Different bodyparts

7.Simultanious targets (nose AND paw)

8.A to B (move from trainer to trainer, two locations)

 

Case study 2: kenneling (cage, box, through a doorway or over a threshold)

• Determine method (targeting, luring, shaping)
• Keep the kennel positive (train it often, keep the door OPEN until the animal feels secure)

1.Work with a partial kennel, build it bit by bit.

2.Give treats inside the kennel, build a high value

3.It is a build-up of trust

4.Keep the balance of reinforcement so that they want to leave the kennel…

 

Case study 3: tactile

• Medical behaviours – require touch
• Desensitization – an ongoing process
• How do we begin touching an animal that does not want to be touched? Develop trust – train target – let the animal touch you – WORK SLOWLY!
• When and where to use it. You don´t HAVE to touch everytime.

Ex 1: Owl – terrified of people. 2,5 years to be able to touch its head, back, talons, etc. Gain the trust, TRAIN! 5 years until everybody on staff was able to touch it.

Ex 2: Too tactile…otters trained to grab something else, a target, the wall.

Ex 3: Tapir, frightened. Desensitization all over the body, the tactile became reinforcing! Massage was very effective.

• Is training always the answer? No. Some species are very skittish and don´t have to be examined very often, and the results may not be worth the time it takes to train and the stress this brings.

 

Handling animal mistakes – positively

Why is this important? Which are the tools and techniques? Popular but aversive techniques under the microscope. Positive alternatives! Using redirection - how?

 

Traditional approach: punish. Punishment works. The very definition of a punisher is that it makes behavior GO AWAY. Why do we want to avoid them? They lead to fear, aggression, and passive animals. But is positive reinforcement as effective? And which are the best PR techniques?

 

“Don´t shoot the dog” is all about getting rid of unwanted behavior. 8 different techniques. Read it!

 

Changing motivation is the long term solution. It is not always possible in the moment. This lecture is about here and now.

 

Positive reinforcement training:

What does that mean? Aversives are NEVER present? Hierarchy of effective procedure (“list of procedures”) by Dr susan Freedman:

1.Medical, nutritional, physical reasons must be examined first of all

2.Antecedant arrangements. Set the environment up for success. Take the temptation away.

3.Positive Reinforcement training

4.Alternative behaviours – redirection

5.Extinction, negative reinforcement, negitive punishment

6.Positive punishment

If you TRY and fail all the steps before you use punishment, you are a Positive Trainer (according to Ken). You should take every opportunity to become a better trainer! Often the reason for punishment is that the trainer is not as skilled in reinforcement methods as she/he could be, or should be.

 

This lecture will focus on 6, 5, 4.

6.No reinforcement Marker (felsignal) is a positive punisher! We want a behavior to end, and we bring something to the situation. We mark the moment the animal does a wrong move. The opposite of a clicker, a conditioned punisher. If you use it WELL, the animal should be HAPPY and NOT FRUSTRATED – but it takes skill!

If it works for you – use it. But don´t overuse it! We tend to fall into a habit of “no”. Selfdiscipline! Pick a special word, that you don´t use every day, to really have to make the choice to use it (no more “I said no by accident, I´m so used to it”)

Deltasignal: “last chance”, warning “if you don´t change behavior I will not reward you. If you correct yourself I will.” Alternative: don´t reinforce that time, try again later – a little easier.

Not now, keep going, good try: are they necessary – or just superstitious for the trainer? Will it really speed up the training?

 

“Try harder-signal”: high energy dog will be pushed harder and be more motivated. BUT it takes a certain dog and a really skilled trainer. Analogy from Ken´s life: he was pushed very hard by his trainer when he trained as an athlete, and took the verbal abuse as “I´m going to do better!” It can work, but is it necessary? Would he have gotten the same results with another – not abusive – trainer? We don´t know.

 

Delta-signal often becomes a cue. Or even THE cue. Delta signal = “…or else”-cue.

 

5.Time-out

“Removal of opportunity for reinforcement” Negative punisher.

Animal exhibits undesired behavior, trainer leaves or turn back to animal. Many trainers use them TOO MUCH! Frequently the animal gets frustrated and does not understand. Timing is everything!

• Is a punisher, and have the same backsides
• Often timed poorly, punishes the wrong behavior
• Does not assist the animal in knowing what it SHOULD do!
• Only works if the animal WANTS to work, WANT you to be there! And then – leaving is just mean…and does NOT tell the animal much information.

Time-outs are effective – in the right conditions! And those conditions are rare. It can in worst cases ruin your relationship with the animal.

Ken avoids – except if the ANIMAL gives YOU a time-out (leaves).

Taking a break for my frustration or my animals frustration is NOT a time-out. It´s a break. End with something simple and easy.

 

5.Negative reinforcement

Removal or avoidance of an aversive stimulus will increase the frequency of a behavior.

1. Avoidance: threat of aversive removed
2. Escape: aversive removed

Exemple: horse and reins. Positive punisher for “going straight”, negative reinforcer for turning. Becomes a cue. Properly used it WORKS! Alarmclock: it rings until we get up, our getting up is negatively reinforced. Yes, we use an aversive, and have to think about what does the dog percieve the most? The punish or the reinforce? NR happen in our world all the time! Can be a very humane and effective choice, if done correctly.

Ex: following collar pressure. The aversive is VERY mild.

Ken prefers negative reinforcement to negative punishment, because your focus is still on getting behavior and not stopping it.

Reinforcement: focus on GETTING behavior – help do the right thing. Punish: focus on STOPPING behavior – punishing the animal.

Overuse can cause frustration. Severity of aversives are hard to control for many. Inappropriate use can be inhumane. Skilled trainers only!

 

5.LRS

Least Reinforcing Stimulus/scenario.

Designed as most positive approach to incorrect behavior. A way to ignore unwanted behavior that cause minimum of frustration.

3-5 second neutral response. Nothing happens.

Goals:

• Step towards extinction
• Effort to provide the least reinforcing consequences without causing frustration
• Give the animal an immediate opportunity to earn reinforcement afterwords

Practical use:

• Brief and immediate
• Looking for calm response (acceptance)
• Immediate, easy opportunity for new reinforcement follows
• As neutral as possible

Challenges:

• Change NOTHING. Nothing to do with posture or eye contact. Pretend it didn´t happen.
• Don´t extend the length, it´s BRIEF
• Refrain from emotional response

Why does it work:

• Due to positive history (the animal understands because we normally click and treat)
• Rhythm of reinforcement is interrupted
• Opportunity to earn is still available – no frustration build

Extinction takes time! You can get extinction bursts, but it is rare with this technique.

Can be used in a shaping session, always follow up with an easy cue. Be aware of patterns! Combine with redirectional techniques.

Default position or station to return to after LRS helps to “reset” the animal

 

4.DRA

Alternate response training: Differential Reinforcement of…Incompatible behavior, Other behavior, Lower intensity behavior. Show the animal what you WANT it to do!

• The heart of most training
• A part of every training – reinforce or not?
• Gives an ALTERNATIVE to the undesired behavior
• Gives trainer something to reinforce

 

DRI:

• Train incompatible behaviour in place of undesired behaviour
• Active form of alternative response training
• Excellent for aggressive training
• Useful for many types of problem behaviours

Ex: Hold on to a target instead of grabbing trainer. Teach a new “begging place”

DRO:

• Reinforce ANY behavior other than unwanted behaviour
• Passive form of alternative response training
• Shape “absence of behavior”
• Useful outside of a session

Ex: Baby elephant in the way gets reinforced for everything that is not “walk next to mummy”, Dolphins get the toys when they do anything that is not aggressive

DRL:

• Reinforce desired change of intensity of behavior
• First used with self-injurious behavior with children (they banged their heads against the wall. If they banged less hard they got reinforced)
• Click to calm (Emma Parson)
• Shaping “down” aggression or intensity, shape away from a behavior

A difficult technique to master, we are still clicking the undesired behavior.

 

 

Simplifying complex training tools

Exploring advanced concepts:

• Require experience to apply
• Training that venture past “reinforce desired behaviour, ignore unwanted behaviour”
• It builds on basic skills! Foundation first.
• Reading about skills and actually having them and applying them, are two different things. It takes PRACTICE to be good at training! Ex: sports (you can know exactly how to play tennis, that does not mean you´re good at it), flying jumbojet (no matter how well you know the theory you are probably going to crash if you don´t train practically).
• The animal has to be ready as well! The animal needs to know the basics and have the skills!

Science vs practical application

• Labaratory is DIFFERENT from “real life”. Adapt to your environment. That means that often we are not talking about the same thing, even if we use the same words…
• Adapt to animals, mix techniques, desired behavior is the goal.
• Science and practice goes hand in hand!
•  

Overview of advanced tools:

• Keep Going Signal. Conditioned reinforcer, tells the animal to go on – click is coming.
• End of session. Not tested scientifically, can be used as an “at ease”-signal. Can cause frustration and danger if the animal wants to go on (especially big cats and killer whales). Can even be punishing. Always end with something simple and a nice reward!
• Jackpot. Magnitude of reinforcer can impact behavior. Has to be used properly to work.
• Recall. Cue for a specific behavior.
• Chained behavior. Each behaviour is used as a reinforcer. String of many behaviours.
• Modifier cues. A type of complex cueing. Right/left/over/under
• Mimicry. Complex cueing.

 

Chained behaviours

Technical or common? Technical: Each behaviour starts the next and reinforces the former. Common chain: sequence of behaviors, bridge after every behaviour, cues are used.

Building a chain:

• Forward-chaining, build a sequence.
• Back-chaining builds a stronger chain.

Train the animal to accept new reinforcers. In a backchain we want the nest behaviour to be the reinforcer. This makes a chain stronger.

Most behaviours are small chains. We build duration and length through approximation.

A well trained chain is NOT a variable schedule, but a reinforcement variety.

Jag uppskattade speciellt hans tankar om hur man tacklar att djuret gör fel. Ken är alltid noga med att betona att detta är HANS åsikter, och vi får gärna tycka och tänka annorlunda. Jag fick mig definitivt en tankeställare gällande felsignaler och negativ förstärkning, nya tankar om hur jag kan använda LRS, och som alltid inspiration och motivation! Flera frågor återstår, men jag hann inte stanna och fråga efteråt (det tog över fem timmar att köra hem i stormen...). Det blir till att formulera ett mail senare i veckan!

Lördagens rundtur startade hos tigrarna:

       

Vi fick se inkallning, ligg (signalen är knytnäven på marken), stanna, och förflyttningar med target.

Vidare till leoparderna, som var betydligt rörligare och mer aggressiva. De tränades i att komma, ligga, stanna, gapa, och följa olika targets (handtarget och platta).

     

All träning med stora kattdjur var från andra sidan staketet, säkerheten först!

Till sist en session med uttrarna, som tränades att gå upp på en våg med hjälp av targetstav.

     

Uttrar verkar väldigt roliga och trevliga!

Tidigt söndag morgon gick jag och min rumskamrat Petra ut i parken i mörkret. Efter att ha lekt "gissa var" och sett ögon speglas i ficklampan stannade vi en stund hos renarna och fascinerades över hur lugna de var trots att vargarna bodde i hägnet mittemot. "Men de är väl långt borta" sa vi och vände oss om...och där stod hela flocken och tittade på oss! Otrolig känsla...

   

Som avslutning fick vi också se glada pandor möta morgonen...en perfekt start på dagen!

 

Jag önskade intensivt att jag orkat släpa med mig stativet, men man gör vad man kan med det man har...

Anteckningar från Kens föreläsning kommer så fort jag får tid!