Picking a puppy from a litter for assistance/service or therapy work

8 months 2 days old

Updated 13th June 2013

This is the third post in a series where I answer some of the common search queries I receive. I speak only reflecting my own experience, in this case temperament testing three litters - two to find Knightley and only one to find Apollo.

While this temperament test is intended for assistance/service puppy prospects, it would work perfectly well for testing a pet, a therapy dog, or a different type of working dog. You certainly wouldn't need to do all the parts of the tests if you were just looking for a pet, and I would only look for a puppy who scored at about the middle of the range - having a super smart dog as a pet can get you into trouble! Feel free to ask questions about adapting the test for pets or other working dogs.

Please note: This test can be used on adult dogs with a few small changes. You may have to skip the first test if the dog is large, you shouldn't clap for the recall test as adult dogs shouldn't need to be attracted like that, the food drive test is even more important as it also tests mouth inhibition (does the adult dog know not to use its teeth on your flesh?), and you'd have to do the webbing test with larger dogs on the ground. Often older dogs are quite touchy with their feet, but it may help to get down on the ground with them. Whether or not they are touchy with their feet shows how much work the previous owner has put into physical desensitisation and grooming.

You may also wish to add to the test a couple of commands - just to place the obedience/training level of the dog. Sit and down would be appropriate, and you could also test how the dog walks on a lead. When choosing an adult dog, having one that has had at least a little training bodes well for its other history.
_______________________________________________________________

It was about three and half years ago I became interested in the idea of an assistance dog (we call service dogs assistance dogs here in Australia), but it took me much longer to research the practicalities of whether it would be feasible for me to have one. I eventually decided I would self train, but then it once again took me a long time to find a breeder, and then a litter and puppy I was happy with. Unfortunately Knightley (my first dog) was found to have severe hip dysplasia, so I had to go through the process all over again to find my second dog, Apollo.

Knightley, happy to lie at my feet and play with toys and get
lots of attention on the day we tested him when he was
7 weeks old. He did extremely well on the temperament test
that day, and the test has been borne out in his behaviour
since.

My experience that was finding the right breeder was important for finding the right puppy. You don't really want a breeder who is themselves quite timid and rarely goes anywhere, who mostly keeps to themselves on their property, as it seems from my experience that they can pass it onto their puppies. Find a breeder who has the puppies inside at least some of the time so from a young age they can get used to different surfaces. Also, it is important that they have people over to socialise with the pups, preferably children and men as even fully trained assistance dogs can be challenged interacting with them. Find a breeder who is likely to produce confident puppies so that you won't waste a trip going to see a timid litter

I have put together my own temperament test, which is a bit of a mixture of the Volhard test (which I see as quite outdated), the PAWS Working dog test and of my own devising, and all given a specific slant for public access work, which I've rather immodestly labeled as the DAD temperament test (downunder assistance dog). I ran it all past some very experienced assistance dog trainers who thought it was right on the mark, and I used it to select both my boys, who have both turned out to have wonderful temperaments, perfectly conjusive to acting as a working dog. You can read about my experience with the first litter I tested here, three puppies all of which I rejected. I wanted a male, and apparently *one* of the females in the litter had a confident, outgoing temperament, but that was not a good outcome for a litter of 6. They were extremely timid pups compared with the second litter of 11 puppies I looked at, 8 of which were male. The final litter I tested (from which Apollo came) was even more outgoing. The breeders put a lot of effort into raising those puppies and it really showed in the temperament test. Apollo was kept with his litter and mother until he was nearly 10 weeks old, and this turned out to be a very good thing. He learnt excellent bite inhibition and didn't ever nip us as a puppy, and learnt to get on with the other adult dogs the breeder owned. However, if the breeder isn't particularly proactive in their socialisation of the litter, it may be worth still bringing the pup home at 8 weeks so you can start the socialisation process. In Apollo's case he was better off remaining with the breeder for an extra couple of weeks.

Knightley doing the retrieve test with the breeder (after I had
done it too), very nicely I might add!

Anyway, here is the temperament test I used. I found sometimes you didn't need to go further than the second or third test with some puppies, because it was completely obvious that the puppy was unsuitable. A very experienced trainer once told me that the purpose of temperament testing was really to discount the puppies you didn't want, rather than the puppies you wanted. I found it quickly became very obvious when a puppy wasn't suitable. Deciding if a puppy was quite good enough is then up to you really, and that's where the scoring can come in handy, especially when comparing two puppies.

If possible do this test when the puppy is 7 weeks old, and get the puppy when it is wide awake as it will give a better performance. Try not to tire it out - if you are keen on it and want to do all the tests you may need to come back on another day. Little puppies need lots of rest!Taking a helper with you is recommended. You will want to prepare a temperament testing kit (see what to take further down the page).

DAD (Downunder Assistance Dog) Temperament Test

1. Lap Test

Lift the puppy onto your lap and simply stroke it. Note its posture and other body language carefully. Does it relax confidently on your lap and enjoy the stroking? Or does it crouch nervously and just shiver? This tests sociability. It is particularly important for Psych Assistance Dogs and therapy dogs.

Great ----- Relaxes on your lap, gives eye contact, curious OR climbs up towards your face to investigate and solicit attention 3 points
Good ----- Relaxes on your lap 2 points 
OK -------- Stands on your lap, loose and unafraid 1 point
Poor ------ Crouches on your lap, may tremble OR struggles to get down 0 points

2. At foot Test

Lift the puppy down to floor level at your feet. Make sure you are in a secure room or in a puppy pen for this one. Continue to pat the puppy as it is by your feet. Once again, watch its body language carefully. This tests sociability, independence and confidence. Once again this test is particularly important for Psych Assistance Dogs and Therapy dogs.

Great ----- Stays by your feet, enjoys the interaction, body loose 3 points
Good ----- Stays by your feet for a while, explores but comes back 2 points
OK -------- Stays briefly, off to explore showing independence 1 point
Poor ------ Moves away from you and cowers in a corner OR stays cowering against you, doesn't enjoy the stroking, body tight, possibly shivering 0 points

3. Recall/Come Test

Get your assistant to move the pup two or three metres (6 to 10 feet) away from you and call it to you - if the breeder is with you, ask what they use to call them for food time. You can clap or do anything else to help get the pup to you.

Great ----- Comes to you with little hesitation 3 points
Good ----- Takes a lot of encouragement, but comes 2 points
OK ------- Watches you, starts towards you, doesn't come whole way 1 point
Poor ------ Completely ignores you 0 points

4. Towel game Test

This is partly as a sight test, partly to test willingness to play which has been linked to trainability. You'll want to put aside or acquire a couple of ratty towels for use on your dog, so use one of these for the test. I tied some wool yarn around one corner of the towel and used the wool to drag and twitch the towel. The idea is to have the puppy in the middle of the floor, and then put the towel in front of him, making it twitch and dart back and forward. This is really testing something called prey drive, how much a dog wants to go after little creatures and toys that resemble creatures in the way they move or how they look. For an assistance dog you don't want a dog with a particularly strong prey drive, but you do want one that is curious.

Great ----- Is curious, immediately grabs it, holds on and tugs 3 points
Good ----- Watches, eventually interacts with encouragement, may hold on 2 points
OK -------- Watches, is at least interested, may want to play 1 point
Poor ------ Is particularly aggressive towards it OR is scared of it, cowering, trembling OR ignores it 0 point

5. Paper ball/small ball retrieve Test

For me, this was an incredibly important test, as one of the main tasks I wanted both Knightley and Apollo to do for me was assistance retrieves. The test seems to have worked - both were pretty good at doing a trained assistance retrieve at the age of only 9 months. For this test I simply get an A4 sheet of paper and screw it up into a ball, then throw it just a short distance directly in front of the puppy. Assuming he goes after it, I then call him, and really encourage him back to me with the ball. If he makes it, I gently take hold of the ball in his mouth and give it a gentle tug, keep up the praise and if he gives it to me I hold a praise party! For the second litter I tested I also had a small nylon ball to use, as well as the paper ball. That worked well too.

The early ability to retrieve has been very strongly linked to successful assistance dogs of all types, as well as retrieve being a necessary skill for most mobility assistance dogs.

Great ----- Will pick up ball, bring it back and let you have it 3 points
Good ----- Will pick up ball but drop it on way back OR will pick up ball but not return to you 2 points
OK ------- Will chase ball but not pick it up 1 point
Poor ------ Watches ball but won't chase OR ignores ball 0 points

6. Squeaky toy interest Test

This test looks at curiosity and is a basic hearing test as well. A dog that is curious is often a good learner. I bought a little puppy toy specifically for this test that was a rope circle, but stuffed with a squeaker inside. It stood the test of time, even after the stuffing and squeaker were gone it remained a favourite toy for a long long time! I would have the pup on the ground and wait until it was looking in another direction then suddenly squeak the toy. Once the pup has reacted, squeak it again. If you are looking for a detection dog, or a police dog, these 'play' tests are very important.

Great ----- Immediately turns, on next squeak comes to investigate 3 points
Good ----- Immediately turns, but takes lots of squeaks to investigate 2 points
OK ------- Turns, upon more squeaks is curious, but doesn't come 1 point
Poor ------ Ignores the squeaking for whatever reason 0 points

7. Tug interest Test

This can be an important test for those wanting to train a mobility assistance work who will need to do lots of tugging (opening doors etc). While there is a lot of difference between a play tug and a controlled trained tug, this will at least give you an idea of the pups willingness to have things inside its mouth and pull at the same time. For this test I used the squeaky toy that was half rope and just right for a little puppy to tug. Very important for detection/police dogs.

Great -----  Immediately interested in playing, will latch on and tug 3 points
Good ----- Pup will hold on, will tug a bit but lets go quickly 2 points
OK ------- Interested in playing but doesn't know quite what to do 1 point
Poor ------ Not interested, ignores it 0 points

8. Hearing sensitivity/audio startle recovery Test

For this test I use two soda cans, and hit the two ends together pretty firmly and loudly. Have the pup on the ground in front of me, and wait until its head turns away from you, then bang the cans together. You are watching firstly to make sure the puppy easily heard the sound, and secondly whether it was afraid and how quickly it recovered from being startled. It's ok if it jumps a bit if it recovers quickly.

Great ----- Turns interestedly, may jump, comes to investigate 3 points
Good ----- Jumps a lot, recovers quickly, returns to what it was doing OR turns briefly (enough that you know the pup heard it), then ignores it 2 points
OK ------- Jumps at noise, slowly recovers, returns to what it was doing 1 point
Poor ------ Jumps at noise, may whimper or hide OR jumps at noise and may have aggressive reaction 0 points

9. Food drive Test

I am sort of assuming your dog is going to be clicker or at least positively trained by putting this test in here. Clicker training is the most common way these days to train complicated behaviour chains for assistance tasks, and it is very useful if the dog is quite food driven! For this test I would recommend using something like a small chunk of raw meat or a soft commercial liver treat - something that is going to be particularly tasty for the little puppy. Hold the treat between two fingers, so the puppy can't get it but can taste it.

Great -----  Anxious to get food, works at it with tongue for a while 3 points
Good ----- Keen to get food, licks but gives up fairly quickly 2 points
OK ------- Notices food, may sniff and lick a bit but not too bothered 1 points
Poor ------ Not interested 0 points

10. Clicker Test

This is partly assuming once again that you may be using a clicker, but it is really just checking aversion to strange noises. The clicker makes a rather high frequency clicking noise that some more sensitive dogs have problems with. Obviously a dog doing public access work has to be ok with all sorts of sounds, so this makes sure that the dog is ok with a sound that it may be hearing a lot of in the future, but also tests acceptance of new strange noises. Hold the clicker behind your back (clickers are very cheap, but if you don't have a clicker, try to find something that will make a fairly loud high pitched noise to test the pups) and click it several times in a row.

Great ----- Pup turns, tries to find where noise is from, not bothered 3 points
Good ----- Looks up, but returns to what it was doing 2 points
OK ------- Ignores the sound (only ok if pup passed the other hearing tests) 1 points
Poor ------ Flinches from the sound, is uncomfortable in some way 0 points

11. 'Weird' public behaviour Test

In the course of their working lives, assistance dogs, therapy dogs and other types of working dogs will come across many different types of people. You will want your assistant to simulate a member of public being upset and getting loud and physical. So get them to throw their arms around a bit and yell, standing in front of the puppy. Watch for its response. This tests reactions to a loud voice, and reactions to strange physical movements.

Great ----- Is interested in what assistant is doing, wags tail 3 points
Good ----- Isn't bothered, may watch a bit 2 points
OK ------- May be a little put off, but once it stops recovers quickly 1 points
Poor ------ Is scared or anxious, flinches from the loud voice or waving arms 0 points

12. Umbrella/visual startle recovery Test

This is once again something that a dog will come across in public a lot, and tests the puppies ability to recover from being visually startled. It is the visual equivalent of banging the soda cans together. It is fine for a puppy to be scared of something, what is important is how quickly the puppy recovers from its fright. Have your assistant stand at the distance of maybe 4 - 5 metres (12 - 15 feet or so) away from the pup (think of them walking towards you on a street, about to open an umbrella), and then very suddenly open the umbrella towards you and the pup. Try to not use an umbrella with a spike on the end anywhere around little puppies - it just isn't very safe. Watch the puppies response when the umbrella is opened.

Great ----- Pup little startled, recovers quickly, wants to inspect umbrella 3 points
Good ----- Pup very startled, recovers quickly OR pup isn't much bothered 2 points-
OK ------- Pup very startled, takes a while to recover 1 points
Poor ------ Pup scared OR pup has aggressive reaction, perhaps barking 0 points

13. Gentle webbing pinch Test

This can be an important test for dogs that are going to be mobility assistance dog, as they are likely to spend a fair bit of time in harness, doing all sorts of tasks that dogs definitely weren't born to do. Some of them may at first be a little uncomfortable for the dog until it gets used to the task, so it helps if the dog isn't particularly sensitive. This is also important for therapy dogs, who may have to be ok with clumsy, uncomfortable or even painful touching from older persons and children. Have the pup on your lap and gently hold one paw in your hand. You may need to be gently persistent as dogs don't like their feet being held. Gently separate two toes and hold the webbing between the toes (like we have at the base of two of our fingers) between thumb and forefinger. Start with just gentle pressure and slowly increase the pressure. You should practice on yourself first until you do it the same every time so that you an accurately compare puppies. Stop as soon as the pup squirms away. Important test for therapy dogs.

Great ----- 7 - 9+ count before puppy squirms/responds to pressure 3 points
Good ----- 5 - 7 count before puppy squirms/responds to pressure  2 points
OK ------- 3 - 5 count before puppy squirms/responds to pressure  1 points
Poor ------ 1 - 3 count before puppy squirms/responds to pressure  0 points

14. Concealed food Test

By ending with this test, you end with something positive and associate yourself with food in the puppies mind! If you do stop the testing at any stage as you think you have enough information on the individual puppy, try not to end it on a more 'negative' test. With this test you want to bring along a clear tupperware type lid, or a very low lying container that the pup will still be able to see through - but make sure it is completely clear so that the pup can see right through it. The idea is to put a hunk of food underneath the lid. Give the pup several pieces of food before very obviously putting a piece on the floor, then putting the lid on top. Then let the puppy do what it wants. This tests persistence and intelligence.

Great ----- Paws and licks at lid, tries to flip it, if gets food is excellent 3 points
Good ----- Briefly paws or mouths at the lid but gives up quickly 2 points
OK ------- Understands food is underneath but doesn't know what to do 1 points
Poor ------ Not interested at all 0 points 

Scoring

As you may not do every test with every dog, what I do is add up the points for the tests you have done for each dog, and then divide by the number of tests you did do, so that you get that puppies average score across the tests.

0 - 0.99 This is definitely not a puppy you want to try training as an assistance/service dog. Even as a pet it would take a lot of work to make it a happy functioning member of your household.

1 - 1.99 This puppy has significant potential and would likely make a good pet. It is probably quite submissive in nature, and would need some work to bring it out of its shell. If its average is in the high 1s, then it may be ok as an assistance dog if you were not planning on doing a great deal of public access work with it - and if it did well on the retrieve and tug type tests. A puppy in the high 1s would make a good therapy dog too.

 2 - 3 This is your target range for the type of puppy you want to train as an assistance/service dog. I would aim for as high as possible, but you also want to make sure that the pup scores highly on the tests that predict trainability as an assistance dog. These dogs would also make a good therapy dog.

To give you an idea, my Knightley scored a 2.7 on his test, and Apollo scored 2.8. Both dogs did/are going very well as ADs, although Apollo is so smart he gets into all sorts of very creative mischief. There were anothers in both Knightley's and Apollo's litters that got somewhere around a 2.5, and indeed in Apollo's litter every single puppy could have become assistance dogs.... they were such sociable and cheerful pups. Knightley's litter tended to score lower. In any litter you test there will likely be a few standouts. However, in the first litter I tested, none tested over 2, and one was below 1. In a case like that you just have to keep searching for the right dog. I would recommend not picking a dog that gets below a 2.0, but that is my personal opinion on the matter.

Summary of what to take

These are just all the bits and pieces I listed in the temperament test:

• Towel with yarn attached to one corner
• Two empty soda cans
• Umbrella
• Meat or puppy treats
• A clicker or something else to make a high pitched short sharp noise
• A squeaky toy
• A toy for tug
• Some pieces of paper to roll up to make paper balls and/or puppy sized ball
• Clear plastic lid/flat container

Keeping track

Knightley after we had chosen him. He was so very tiny!

You are going to want to keep some record of how the pups did, especially if it's a big litter! I didn't bother to start recording unless the pups did well on the first couple of tests, because if they were getting zeros and ones on those first tests they definitely wouldn't be doing well on the later ones. Try to get your assistant to do the keeping track as otherwise it's a lot to do!!

By the time I had tested the seven boys in Knightley's litter my head was spinning.... but Knightley was the last one, and it was clear from the outset that Knightley was going to be the puppy for us, without having to add up scores or anything. This was the same with Apollo, he was the clear standout of the boys, although there was a girl who was even more confident and trainable. I had never tested any girls before, and did for the breeder who had been asked to supply a suitable girl for a family with an autistic child. The one I recommended was a real darling.

Final points to consider

• Start with the right breeder, look at health clearances, try to view the sire and dam, make sure the puppies are well socialised from a young age as that will make a huge difference throughout their whole life. The more effort the breeder puts into those puppies, the less work you will have to do
• Try to find lines which have had dogs as assistance/hearing/guide/therapy dogs, or at least have participated in obedience
• Don't let your heart decide which puppy to go for, it may well mean rehoming when it grows up to be an unsuitable dog. There is no such thing as 'close enough' when choosing a potential AD or even a family pet
• While temperament testing is a very valuable tool for choosing a puppy, remember you are primarily trying to weed out the puppies that are not suitable. After then it is your choice
• Even though a temperament test means you start out with the best puppy possible, it is no guarantee that it will make it as an assistance dog. Dogs go through two main personality changes throughout their lives, so you can never be sure how a puppy will turn out. Testing just makes sure you start with the best source material possible.

If you have any questions, feel free to comment or contact me on my Facebook page https://www.facebook.com/DownunderAssistanceDog