Killing the towel: Temperament testing

Since Kate just got the puppies back from being temperament tested, I thought I’d try to explain a little bit about how temperament testing works and what it’s trying to test.

The first, and most important, thing to understand is that TEMPERAMENT TESTING SHOULD REVEAL NOTHING. That is, it should confirm what you already know about the puppy or it should help you organize your thoughts about the puppy, but if the puppy tests completely differently than you’ve ever seen him or her act, something went wrong and you need to either re-do the test or get a new handler or let everybody take a nap and calm down.

Second, TEMPERAMENT TESTING IS NOT GOD. This goes along with why I have begun to hate temperament labels for dogs. It is very tempting and COMPLETELY WRONG to walk away from a temperament test thinking that you now know that X puppy “is” something. Is prey driven, is shy, is soft, is hard, is whatever. What you saw in the testing is simply the choices the puppy made [using all the tools it had available to it at the time] [under the conditions it had lived in to that point] and [with the resources it had inside it]. If it does not have good tools – if it’s been undersocialized, underexposed to dogs, underexercised, etc., it cannot perform at its best. If it has not been conditioned – if it has never had to face some of the situations the test presents – it will seem to be much more shy/startled/afraid than other puppies. For example, a puppy who has had a baby agility teeter in its exercise pen since it was five weeks old is going to run over and want to play with it. That dog is not necessarily a better pick than a puppy who is seeing this weird blue and yellow thing for the first time in its life and runs for the hills when the teeter thumps. You also have to look at resources – was the puppy overtired? Hungry? Overstimulated or understimulated? Just woke up? Jealous because it had watched the tester work with all the other puppies before it? That is going to skew the results.

That’s why the best way to think about temperament testing is as a way to organize what you were already thinking about that puppy, to be sure you have thought about all the different ways this puppy operates, and to do it while the puppy is handled by a stranger. I think that last bit is actually what makes temperament testing useful – because you are not seeing the puppy acting a certain way because he or she loves you so much. You have to see how it works under a more stressful situation with a stranger in a strange area. 

Basically, temperament testing is designed to identify DRIVES and SENSITIVITIES.


are basically a way of talking about how eager the dog is to perform a behavior and how much effort the dog is willing to put into achieving its goal in the behavior. Some of them (not an exhaustive list by any means) include:

Prey (which has subdrives including dig, track, herd, kill, shake, etc., and some dogs are very developed in only one of these subdrives)


Pull/tug (some people classify this with prey or play, but I think it’s separate)






You need to understand that EVERY dog has EVERY drive. This is basically just a list of “stuff dogs like.” The question is whether you’ve got a dog who, from day one, can be called off a fish stick with a raised eyebrow (low food drive) or a dog who will go bananas, offer you seventy-five behaviors, and then leap over and try to swallow your arm if you give him a cheerio (high food drive). 

Also, just like in humans, a dog can decide or be trained to not respond to his drives. You may reaaaaaaally like food, consider that a very high priority in your life, spend a ton of time and money in restaurants, take cooking courses, and you still do not bring a giant sandwich into a formal board meeting. At that board meeting you may be the person in the room that pays the closest attention to the PowerPoint slide that pictures a pizza, but you don’t leap up at the screen and tear it down and try to eat it. In the same way, a dog who has strong drives should never be a slave to them and you should never use that as an excuse to ignore or accept unwanted behaviors. Your goal with your cuckoo-for-cheerios dog should be the ability to call him off with an eyebrow. 

The opposite end of drives is


These are basically a list of things dogs DON’T like or that make them feel odd or that put emotional or psychosocial pressure on them. A very non-exhaustive list would be








A dog who is “sensitive” to one of the above list is a dog who finds the feelings or pressure SO unpleasant that they work very hard to excape it or avoid it, even to the point of using his or her mouth to signal how strongly it feels. 

Again (and you had to know this was coming), I don’t want anyone using “my dog is sound-sensitive” as an excuse for a dog who has to be sent over to the neighbor’s house if you want to blow-dry your hair. Sensitivities can be flattened or healed or even completely reversed – a gunshot is objectively unpleasant for every dog on earth, but hunting dogs learn to associate it with their favorite activity. They quickly move from disliking the sound to loving it, just like children move from being afraid of the sound of loud music to spending $75 of their parents’ hard-earned money to go stand next to the speakers.

OK, now that you’ve had that introduction to concepts, what exactly does the temperament testing involve?

It totally depends on the testing.

The most famous of the general pet-puppy tests is called the Puppy Aptitude Test and was developed by the Volhards many years ago. This test has been done by a ton of breeders on zillions of puppies and tends to be what show breeders are talking about if they say the puppies are being temperament tested.

The specifics of the test can be found here (and many other places):

The basic idea of the PAT is that the puppy is tested for desire to follow a stranger (which could be defined as social drive or pack drive), for reaction to being picked up and restrained, for loud noises, etc. You want the puppy that reacts in a middle-range way, happy but not over the top, likes people but isn’t going crazy, etc. The “best” puppy is the one who scores mostly or all “3” on the tests.

When I first got into dogs seriously, the dog performance world was beginning what has seemed to me to be a meteoric rise. I didn’t know anyone who did agility but I knew it existed. The breeders who were talking about getting a title on both ends of the dog meant obedience. 

And then came agility. And then flyball, and I started hearing about all these people doing freestyle and “tricks,” then disc, then dockdog, then rally, and WOW. There are so many more things to do and so much more that buyers (and owners and breeders) expect to be involved in.

This is a GREAT thing. I could not be more thrilled, honestly. And I think most breeders feel the same.

Where this began to impact what I understood about temperament testing was that the performance people I knew were no longer looking for that quiet, focusable puppy who never overreacted to anything. Those puppies were wonderful for competitive obedience, which is based on the ability of the dog to obey with great predictability and control. But the “new math” of dog performance relied on the dog making its own decisions and working from its own desires. Essentially, if you put it in figure-skating terms, these dogs need to move from the “compulsory figures” (the refinement of extremely consistent and predictable movements) to the rock-and-roll long program. The breeders, owners, and buyers looking for dogs who would excel at these disciplines found that they couldn’t get the results they wanted from those middle-of-the-road puppies.

The really smart thing that they did to begin to refine the puppy selection process was to seek out masters of the disciplines that had always required independent thinking, desire, and creativity on the part of the dog. They asked hunting breeders and schutzhund breeders and tracking breeders how they picked puppies. And the result of that big swirl of thought was something like the following:

What I want you to do is look closely at how very similar the PAT and PAWS tests are (looking for following, looking for the reaction to a dragged object, looking for behavior when asked to be still), and how COMPLETELY differently the results are interpreted. A puppy who scores all “1” in the PAT test is described as practically a killer in waiting… “This dog is extremely dominant and has aggressive tendencies. It is quick to bite and is generally considered not good with children or the elderly. When combined with a 1 or 2 in touch sensitivity, will be a difficult dog to train. Not a dog for the in experienced handler; takes a competent trainer to establish leadership. “

WOW, I don’t want to have anything to do with that puppy!

But those exact same actions on the part of the puppy are across-the-board “EXCELLENT!” in the PAWS working dog test. 

Interesting, isn’t it?

I don’t want anyone to think that I don’t believe in the PAT test or that everybody should switch to PAWS or anything like that. I think they’re actually very complementary (do the PAT first, then move the high-scoring puppies into the PAWS test) and both valuable. But I DO think that if you intend to give your dog a job, there’s absolutely no reason to be afraid of those scores of “1” and “2” on the PAT, even though the wording on the bottom of the test form is so very, very scary.

This post is getting way too long, and I want to leave room in other posts to talk about the different drives and sensitivities and so on. What I want to get down to is the BOTTOM LINE for puppy buyers.

From my point of view – and keep in mind who I am as a person – my perfect puppy is without a doubt going to be up there in the 1 and 2 scores. And I have tiny kids and disabled and elderly and every other kind of person visiting, and I have rescues and visiting dogs and intact dogs and bitches and crazy dog-related activity in my house constantly. I love, love, LOVE a confident, pushy puppy with a brain. I find it exponentially easier to put the brakes on a happy puppy than try to jolly along a quiet or worried puppy. And I’ve never seen any connection between high prey/tug/play/ball drive and behavior with kids. The first time one of my puppies puts a paw on a baby, the world ENDS for that puppy (please don’t worry – I do not ever hurt or even physically contact the puppy, but it knows that it is never, ever allowed to do that), and generally that’s the last time they try it. 

So I want the puppy that kills the towel and kills it dead and then prances around with it. I want the one who barrels over, thumps into my lap, jumps up and licks my face, then gets off and starts tugging my pant leg. And when I take the pant leg back, I want him to bounce back and go for it again. And when I walk, I want him herding my feet and biting my shoes the whole time. I’m going to take that puppy home and we’ll live happily ever after. 

If you are not me, if you don’t love a ballsy puppy, if you don’t want the star of the show, if you want less brain and more couch potato, go for the more typical PAT “3” and “4” results. There’s going to be very little about that dog that will be a problem. You’re not off with a free pass, of course – every puppy needs socialization and training – but these puppies tend to be very normal, puppyish puppies.

The puppy who scores in the 5 range in the PAT is going to have some strengths and some weaknesses. I actually disagree with Volhard about putting that puppy with multiple, especially loud and active, kids or a whole bunch of dogs. I think a puppy who is very soft tends to be overwhemed by the chaos of a very busy household. But WOW can these puppies be absolutely effortless to raise. They never do anything wrong, they are not destructive, they don’t challenge you, they love to just be with you, they tend to bond very tightly with you and seek you out. A lot of people find that the mostly-5s dog is the easiest dog they’ve ever raised or been with. The challenge with a soft puppy is to remember that you still have to socialize the heck out of them and still have to train them, and you have to find the keys of building drive and confidence in that puppy. Where a soft puppy goes “bad” is when it isn’t given the tools to feel like it can handle stress. A pushy puppy is going to just “fake it till they make it,” and thrives in new or unusual places. The soft puppy needs to have those tools firmly in place so it knows exactly what to do and where to look when it is stressed. And they are not (usually) ever going to be completely comfortable out there in stressful new situations. They love being at home, which is where they excel. 

The short story? There’s really no such thing as a “bad” temperament test result in a puppy, with a very, very few exceptions (a puppy who just freezes and is catatonic when it meets a new person and sits there in a puddle of its own pee would REALLY worry me). Using temperament tests helps breeders know which puppies are most likely to excel in which homes, or prepares a buyer to meet the needs of that puppy. And it can help a buyer like me, for which a really soft puppy would be a disaster because (let’s face it) my family would scare it to death, look for a puppy who will thrive.

Also, realize what drives and sensitivities neither test measures. In neither test is food presented to the puppy; in neither test is the puppy invited to show its instinctive feeling of where it belongs in the ranking of the pack. Neither test shows whether the puppy can tell the difference between an adult and a child (most dogs can, and can modulate their behavior to a startling degree to take into account the relative weakness of the child). Neither one adequately tests for how the puppy reacts to being solitary. And so on. If those skills are important to you, you need to make sure you devise ways to test for them, and make sure you weight them appropriately. If you really need a puppy who is good with the elderly, pick the one that’s good with the elderly! Don’t reject that puppy because of his score on noise sensitivity, or accept another puppy who got all “3” scores but freaks out and panics in the nursing home. This is supposed to be a common-sense procedure, so be sure to keep it in that area of your brain.

I also like to see a jealous puppy, which Volhard would probably have a heart attack about. I will look hard at a puppy who is up against an x-pen yelling at me because I’m petting her littermate outside and who then goes into spasms of joy when I pick her up. I like that because I love a dog who says “ME! Look at ME! I want to be with YOU!” and really does mean it. A dog who is standoffish is much harder for me to reach. But you may be just the opposite – you may want a dog who will quietly observe but not stick her head into every hand and leap against every leg.

I’ve thrown a bunch of information at you, but I go away let me focus your attention on what I think is a SUPER SUPER KEY aspect to puppy personality that is NOT well covered in these tests, although an experienced tester can usually read between the lines and see it. I’ve saved it for last because I want to emphasize its importance.

That aspect is FORGIVENESS.

Forgiveness in a puppy is how it reacts to being slightly hurt, or insulted, or confused, or deprived of something it wants. An unforgiving puppy reacts to unpleasantness by stopping, looking away, avoiding the person who caused the bad feeling. A forgiving puppy says “Wow, you did something that strikes me as really dumb, but I love you anyway!” 

Since I personally have a household in which things crash and fall and puppies get stepped on and eyes get poked and so on, I want a puppy who reacts to the insulting parts of the test – being rolled over, being picked up, the paw, and even the elements of the other tests that are no fun (the part where the towel tug gets taken away from the puppy) by pausing for a second and then going right back up and trying to lick the tester’s face or tug a sock or whatever. For me, that’s actually more informative than how the puppy reacted during that test (what score it got). I want that puppy’s desire to be with the person to overwhelm the insult, and for the puppy to instinctively look to bond again with the person that insulted her. But a good friend of mine who works her dogs actually prefers an unforgiving dog, because unforgiving dogs keep her precise as a handler. She has to give every signal perfectly and expect the same thing every time, and if she gets sloppy her dog will just stop and stare at her, or even walk away. They school her just as much as she schools them, and she loves that amount of brain and the high expectations her dogs have. Just like with the other tests, this is not a good/bad result, but it is a VERY useful one. 



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s