Assessing “Assess-a-Hands”

When I began working as an animal behaviorist I evaluated dogs in a number of ways, one being to give them a prized resource and watch to see how they responded when I reached for it. I remember working on a stage with a dog who was said to bite if you tried to take away his toy. After playing around with him to get him comfortable, I gave him his treasured toy and let him settle down with it. Then I squatted down beside him and began to reach toward the object with my hand. I explained to the audience that I wanted to see not so much whether he would growl or try to bite, but the expression on his face as he reacted to my outreached arm. Was his commissure forward in an offensive threat? Or backward in a fearful grimace?

Over a hundred people held their breath as I reached closer and closer to the dog. I was riveted to his face, fascinated by how he might behave. I felt confident that I would see signs on his body or face that warned me he was getting close to the edge long before an actual bite. True enough, as my hand got within inches of the toy, his body went stiff and still, and the corners of his mouth moved forward in warning. “Got it,” I said, and withdrew my hand. I rarely repeated this procedure, not wanting to teach the dog to growl or snap, but I did this one test on a great many dogs. I don’t know many individuals I evaluated this way, but it was in the hundreds, if not thousands.

Years later, Sue Sternberg introduced us all to the “Assess-a-Hand.” (Let’s call it an AAH.) I watched her use it once, and never looked back. It’s strange, I don’t remember ever being frightened of being bitten in those early years, but as soon as I had another alternative I jumped onto it like a flea in a circus. Since then it has become a standard piece of equipment in behavioral evaluations, but a controversial one. Someone recently asked me how I feel about them, and it seemed a good time to add my voice to the mix. Here’s the bottom line: AAHs can be great, and they can be awful, depending on how one is using them. It depends. (Heard that from me before?)

The good: Assess-a-Hands can indeed be a great tool for evaluating whether a dog resource guards or not. This is valuable information to know, for anyone deciding where best to place a dog, and/or what kind of work a dog needs to be safe and happy in anyone’s home. I’m not convinced that it matters whether the dog perceives the hand as “real” or not, as long as the dog perceives the hand as something that he has to guard against. I can tell you, from using it in so very many cases, that dogs vary tremendously in their response. Some dogs lick the hand, some ignore it, some start eating faster, some go stiff and still, some go stiff and begin to growl, some snap at it. A few dogs, without any warning growl whatsoever, latch onto it and shake it in a fury. I worked with one dog who ignored the hand and lept up in an attempt to bite my arm, this after I merely moved the hand to within a few feet of his toy. If I recall correctly, the dog was being considered for adoption by a family that included four or five children, several of them under the age of six.  Needless to say, I suggested that they might choose another candidate. Stories like that suggest that AAHs  are tools that can do a lot of good.

The Bad: Anything with the power to do good usually has the reciprocal power to do harm. That is as true here as it is with scalpels or medicines. I’ve seen an AAH used to harass a dog far beyond the point of reason. I’ve seen dogs approached so often with an AAH that one wants to yell “#%^*$#, Leave The Dog Alone!” Evaluations are not excuses to harass a dog. I should mention here that one can “evaluate” a dog without an Assess-a-Hand and still cause problems. One client came to me with a dog who had been “evaluated” by someone who tied it up, and then approached it in a stalking posture, with a baseball bat held up in the air as a threat. As the “evaluator” got closer he began to yell at the dog. When the dog attempted in desperation to escape, the man continued going after him. The dog finally began to growl when it was completely trapped, and was deemed “aggressive.” Euthanasia was suggested. Good grief.

The Ugly: There’s more. If one pushes an AAH repeatedly toward a dog, withdrawing every time one gets a reaction, it is relatively easy to teach a dog that growling is the ONLY way to be left alone while eating. Exactly the wrong thing to teach a dog; we should be teaching the opposite. That is why there needs to be a carefully followed protocol that avoids random swipes with the hand regardless of the dog’s reaction, and clear boundaries about how and when the hand is presented. Notice how the hand is presented only twice in this video of the SAFER test used by many shelters, including the ASPCA.

One last comment, because the use of the hand brings up a good question that deserves an answer. Is any test 100% predictive of how a dog in one context will behave in another? Nope. Just ask my friends how I behave when I’m working, compared to how I behave after I have a martini. But there is data that evaluations, if done carefully and consistently, can increase the probability of predicting how a dog will behave in another setting. It is always good to have a lot of tools in one’s tool box, and an AAH is a good one to have, as long as it is used correctly.

If you’d like more information, here’s an article by Dr. Emily Weiss at the ASPCA about this very topic, and here’s a delightfully funny video about how a person might react if tested in a similar way. This will stop you from trying to temperament test someone who is eating a slice of pizza.

MEANWHILE, back on the farm: It’s melting! Slowly, gradually, not very much, but still, we have achieved meltage. Right now there is still a lot of snow on the ground, but one can see soil in some place, and it’s not ten below in the morning and is actually going to stay above freezing a few days out of the week. T-shirt weather!

The cats are out of their cozy igloo, the one lined with a heating pad, and playing among the pines. The sheep are wending their way up the hill (still snowy and icy), optimistically looking for something besides hay to eat. They’ll have to be patient.

I have one more Contemplative Photograph class this session, although I have to admit to not finding any inspiration outside yet.  Snow. Ice. Brown. Grey. White. Yada Yada. I probably should be more creative, but I just couldn’t get inspired out there. So I turned my focus inward. Here are some inside shots I took of, wait for it….. the dogs! Imagine that.

Willie Close Sm



Tootsie close Sm




  1. EmilyS says

    Studies have indicated that these shelter tests are not good predictors of how a dog will behave in a home.

    These tests have been used as the reason to kill thousands of shelter dogs as “aggressive”.

  2. Mary GM says

    I have another possible caveat to using the hand:
    My adopted dog is afraid of all stick and rod-like objects, so I wonder how she reacted to this test if it was given to her. We have no background info on her, so we don’t know if she was abused to cause this fear. She probably would have run away, like she does if I carry a broom, curtain rod, etc. She will still resource guard a bit- a sharp warning bark if the cat comes too close to her treat, so she only gets treats with supervision.

    Overall, it sounds like a great tool, but maybe it could be used in conjunction with another test?

  3. Susan Fishbein says

    Great article!! I see so many people abuse the AAH and torture the dog in the name of an assessment. My last foster dog was saved by a breed-specific rescue organization when he was scheduled for euthanasia because on the FIFTH time they poked him in the side with the AAH while he was eating, he growled. Made me wonder if there would have been a SIXTH time, if he hadn’t growled on the fifth. Besides, poking in the side is not the protocol for using the AAH! However, is assessing whether a dog is going to lash out when a person (i.e. toddler) goes near him when he has a resource, I’d much prefer using a rubber hand on a stick than my flesh and blood hand. I’ve grown rather fond of all of my digits.

  4. Jan Hankins says

    Thanks for this topic and for pointing out the good and the bad about AAH. My personal experience with the AAH has not been positive, as we’ve taken in several dogs that “failed” the food aggression portion of the SAFER test (using the AAH). In all cases, the dog was not food aggressive. While I realize the usefulness of the AAH (I do not advocate “sticking your hand in the bowl and if you get bitten, the dog is food aggressive” as a method of testing for food aggression!), but I would like to know that people use it properly when decisions about euthanizing a dog are being made based on it. I would also like to know that people who use the AAH (and other “props” used in the SAFER test) introduce those to the dog and desensitize the dog to them before the test. Perhaps it isn’t food aggression, but fear of the AAH? Perhaps it isn’t that the dog doesn’t like children, he just doesn’t like an adult chasing him around the room with a large baby doll and speaking to him in a falsetto tone of voice?

  5. Joe says

    IMHO, the AAH subject was incredibly patient with the experimenter. I have dealt with subjects who would have grabbed the AAH and broken it over the tester’s head, on perhaps the second try. I have also dealt with ungrateful subjects who, when given a slice of pizza, ask “Where’s the beer to go with it?”

    Seriously–A great demonstration of how NOT to use an AAH.

    I’ll have to get a pic of dog, outside in the melty snow to share…

  6. says

    But isn’t it Sternberg’s position that if the dog can be made to fail, it should be–and then killed because it’s obviously not safe and there are plenty of other dogs?

    And I’ve seen far too many videos where the dog is indeed harassed persistently until it finally lashes out at the thing that looks like, but doesn’t smell or move like, a human hand. It also gets used on dogs newly admitted in emaciated condition, without giving the dog a chance to get used to regular meals.

    Yes, it’s a tool, and like any tool can be used well or badly. I’m deeply suspicious of overconfidence in it.

  7. Amber says

    Hunh. The dog in the SAFER video was so NICE about having food taken away, too bad he didn’t get the reward of getting it back to finish it off! I actually much prefer the Center for Shelter Dogs Match-Up II protocol for assessments – most of it is off-leash which I think really allows you to see the choices a dog makes when given opportunity. The poor dog in this SAFER video is so confused by all the strange, back & forth movements and mixed signals given by the tester and all the leash-wrangling that his friendly, wiggly side is almost completely overshadowed with insecurity! The CSD protocol also includes introducing the dog to the AAH before bringing out the food as well as more opportunity to look for subtle signs before actually diving the hand into the bowl.
    Thanks for your insights on this topic Patricia! Those who flat-out dismiss the AAH have very often only seen it used badly, and generally have not had to assess dogs with completely unknown histories… We like the option to keep our own hands! :)

  8. Trisha says

    To Emily S: What studies are those? I’ve seen the opposite (Emily Weiss’s PhD for example). Would love to see any other data anyone has; you all know I’m a sponge for data!

    To Lis: No, that is not Sue’s position at all. Check out what she says, not what people say about her.

    To Joe: Love the suggestion to ask for a beer with the pizza slice. Wasn’t that video funny? I loved how the ‘subject’ kept cracking up herself.

    Thanks to those of you who appreciate my comments about using AAHs correctly. It is a great shame that people abuse them. But as I wrote, regrettably, people have been doing abusive evaluations for years, the AAH is just another tool by which to do that. I should add that if I had lots of time to spend with a dog I’d never consider using one. But, if I was in a shelter setting, or evaluating a client’s dog with little time, they can indeed be helpful, IF IF IF used carefully and judiciously.

  9. Pam McQuade says

    I have always been leary of such tests because I believe that they are too easily used to euthanize dogs. My first rescued basset came to me with food aggression issues. He had not had an easy life before he walked in my door, but he connected immediately and thoroughly with me. I slowly and carefully dealt with the issue, and in a short time he became a different dog (though I was always careful with him). He was always tempted to dispute food with a new dog lower in the pack order than he was, but eventually he learned that I was always rigorously fair about giving all dogs treats at the same time, and even this aggression disappeared. I believe it was all a matter of security.

    I am certain that if he’d been in many shelters, he would have been euthanized. That would have been a great waste of athis wonderful dog who became my heart dog. I am certain many great dogs lose their lives because people are not willing to spend the time and energy it takes to allow them to trust. Naturally an adopter of a dog with aggression needs to be chosen very carefully, but at least some dogs who respond to a hand in their bowls do not deserve to lose their lives. It makes me angry to think that this is a knee-jerk reaction in so many places.

    If I were starving and you took food from me, I’d probably react less than nicely. I’m sure it’s the same with dogs.

  10. Linda Watkins says

    In the interest of keeping life fun, I’ve chosen to rescue Australian Cattle Dogs – not a breed known to do well in shelter environments to begin with – and so far, every dog I’ve pulled who “failed” the AAH test, has been a shelter fail only.

    I watched a video of one dog chained, hungy, wolfing down food while the AAH person made high squeaky, nervous “gonna get the dish” noises and then flopped the hand around several times in his face prior to actually pulling the dish away. The dog chased after the dish, but of course hit the end of the teather – so he kind of shrugged, and went back to the wall.

    The dish was pushed back to him, he waited a full minute, and then figured it was going to stay in place, so returned to eat. The “gonna get it!!!” squeaks began again and the second the hand wiggled w/in a few inches of his face, he struck – grabbed the hand off the stick, tossed it aside and returned to the dish to finish his meal.

    He was deemed resource aggressive. Had I been in the same situation, I too would have been deemed aggressive. I took this dog into rescue, did some resource testing on him w/ zero issues; he’s in a home with a couple of senior humans whose grandchildren visit on a regular basis – there has NEVER been an incident or issue with resource guarding.

    The problem, I suspect is not so much with the dogs, as it is with those who have not been trained to properly administer the test. I’ve yet to see a shelter situation where the test results gave a true reading – at least for the dogs I work with.

  11. Trisha says

    Thanks Emily, much appreciated. Dr. Marder has done a lot of good work and is a credible resource. I agree completely that all factors must be considered, and, as I’ve written repeatedly, no one test in one environment is ever predictive of a behavior in another environment. The different results between Weiss and Marder are reflective of what happens often in science: One study says one thing, one says another. But none of that changes the primary point of my article, which is that if ones uses an AAH, they should only do with tremendous caution.

  12. HFR says

    I think the underlying issue is what to do about devices/instruments/theories that can do great good if used correctly but can be easily abused by those using them. Guns, anyone? Alchohol? Pain-killing opioid drugs? ADHD drugs for children? Any drug, really. Electric fences, shock collars, choke collars, muzzles, head halters. Where is the formula that tells us when the abuses outweighs the benefits? For instance, the negatives outweigh the benefits with shock collars. No doubt in cases where the dog could lose its life (as in teaching a dog to avoid snakes) they may do some good, but, I think most of us would agree, they are much more likely to do damage than good.

    I think what happens is that once a device/idea is put in the wrong hands, it somehow becomes desirable by others who like to abuse power. Same is true with the dominance theory of behavior training. Once that theory was adopted by those who like to control animals, it became almost a religion that, to this day, science is still trying to negate years after it has been disproven. The people who continue to poke a defenseless dog with an AAH to get a reaction, WANT that reaction. They then feel powerful and in control and useful. They have saved the world from a dangerous dog. They are in charge.

    While using your own hand to test a dog is dangerous in itself, it probably kept the people who would abuse the test from trying it themselves. Once it was made “safe” for all to do, it only stands to reason that ignorant people would abuse it.

    Not sure what the solution is, but it’s a universal social problem for more than just the dog.

  13. Joyce says

    I agree wholeheartedly with everything you said, but this in particular is why I like you so much “Nope. Just ask my friends how I behave when I’m working, compared to how I behave after I have a martini.” Mmmm. martinis :)

  14. Diana R says

    Thank you for writing this article, Trisha. Personality and behavioural assessments of dogs is the topic of my PhD (2/3 of the way there – yay!), so this is a subject close to my heart.
    Completely agree with what you’ve written about the use of the AAH. I’ve collected data on 180 dogs tested using the SAFER method and we only had 3 snap at the hand (all on the first subtest too) – all with histories of resource guarding towards people. The hand CAN be used well, but, like any assessment, the person involved must be fluent in canine body language and attentive (there is NEVER an excuse for harassing a dog or setting it up to fail!). It’s also common knowledge (or it should be), that even when well used, the hand test has a high false positive rate, which is why Sue S, Amy M and Emily W all suggested behaviour mod and retesting for any dog that fails but does ‘well’ otherwise. Several of the comments above about failing dogs include significant failures of the tester to adhere to protocol – which is another huge problem, especially in shelter tests where limited time and resources are a given.

    I’m specifically looking whether heart rate monitors can increase the validity and reliabilty of tests and revisiting a number of subtests (including the doll), to see which if any behaviours they predict. Could write on this all day, but I’m now late for work so I’ll stop, but THANK YOU Trisha for the blog article – it put a big smile on my face to start the day!

  15. LisaW says

    One question comes to my mind. Since many dogs in shelters often have issues of some kind or another or some upset, insecurity in their lives (often just being in the shelter is an issue), why test them using tools created more or less for bomb-proof dogs? I assume the result most testers are after is a calm response or no response, and I only know a few dogs who would provide those types of responses, it almost seems like the tool is not quite right and even less appropriate if the tester is not trained or has a mean streak. It’s analogous to giving someone with anxiety an anxiety-producing stimuli that would have no affect on a calmer person and then saying, see you’re anxious.

    I think only one of my dogs in the past thirty years would have passed the AAH test. Not because the others were guarders or aggressive but they would have reacted to a stick with a hand being poked in their face, some more dramatically than others. I could move their food dish or add something to it while they were eating, etc. They would not have reacted to my human hand (even a stranger’s hand) but would have reacted to a dummy hand.

  16. Trisha says

    Diana R, thanks so much for writing in. I’m thrilled that you are doing your PhD on beh’l evaluations, promise to keep us in the loop? Can’t wait to hear your results; but love hearing what you are seeing so far.

    To LisaW: I’m not so sure that many dogs in shelters have “issues,” at least not beyond “not coming when called” when they’ve actually never been properly trained. I do get your point though, but see Diana R’s research above, and that she has only seen 3 out of 180 dogs snap at the AAH. That is close to my experience; most dogs do what most of us would do–move slightly away and keep eating. Every once and a while a dog would stop what it was doing and look me in the eye. Not in a hard, challenging way, but in a way that I could only interpret as “Excuse me, could I help you? I’m a tad confused because, as you can see, I’m busy here eating my dinner, but if you’d be so kind as to wait a moment, I’ll be right with you.” Anthropomorphic I know, but sometimes that’s the only way we can try to imagine what they are attempting to communicate.

  17. says

    Although I am only going to reiterate what everyone has already said, I wanted to add our personal experiences as the intake people for a regional Jack Russell terrier rescue. Since 2007, every single shelter dog who had failed the SAFER test and was then passed to us as a result had been slated for euthanasia BECAUSE they had failed the test, and every single one had only failed the AAH component of the SAFER test. I have had several shelter staff in several different shelters in 2 states tell me that the dogs in question were totally fine with shelter staff sticking their hands in their dishes while they were eating – why are we so rude to dogs anyway ? but shelter rules required any dog who failed ANY component of the SAFER test to be euthanized or given to a breed-specific rescue. So we were lucky to get some very lucky dogs who ended up being adopted by wonderful terrier-smart people and living long contented lives. NONE of these terriers were guilty of any excessive or dangerous resource-guarding. Thus – in OUR (admittedly anecdotal) experience, the AAH test is both confrontational and inaccurate.

  18. EmilySHS says

    I have used the Assess-A-Hand in behavior evaluations for hundreds of shelter dogs over the last 8-9 years and I myself find it an incredibly useful tool. (One thing I’ll put in quick–we don’t consider resource guarding in and of itself to be grounds for euthanasia at our shelter–the tests are just a place to start, not a determination of fate.)

    Curious things I’ve learned from my regrettably informal collection of data: roughly 22-25% of adult dogs show guarding behavior as defined by my lowest criteria–freezing–but in one sample of 56 adult dogs, only 8-10 escalated to growling or tooth displays and only 2 snapped or bit. I have never seen what I felt (very mushy and non-scientific, sorry) was any correlation between being under-weight and guarding–very thin dogs don’t seem to guard any more or less than the general population. On the other hand, I think I have seen a tendency for obese dogs to be more likely to guard. And when we used to test litters of puppies, I pretty much always got the same thing: if one puppy guarded, they all guarded; if one puppy didn’t, none of them did. Which has led me to suspect (research, anyone, please?) that guarding behavior is at least partly a product of litter-mate “culture”–and possibly retained via learning by some adults and decreased or eliminated in others…?

    Lastly, I had an opportunity at a conference years ago to buttonhole a behaviorist with this question: allowing for the fact that the fake hand is an artificially “hot” trigger, if we do get an aggressive response or bite, is it likely that we are seeing the dog’s real style of warnings and/or level of bite inhibition? So if the dog’s bite to the fake hand is a Level 1 or 2 “nip” (as far as we can tell), or if the bite is a multiple bite where the dog grabs with enough force to yank the hand away and shake it–is that likely to be an accurate reflection of the dog’s “mouth”? He felt it probably was–that even if the AAH is more likely to provoke aggression, the response provoked may very well be a true reflection of the dog’s pattern when aggressing. And golly, I would so love to hear any thoughts and research on that…

  19. liz says

    Is a stuffed stimulus dog somewhat analogous to an AAH? When used appropriately, does this additional tool indicate potential dog aggression as effectively as the hand indicates potential resource guarding? Reading of the upsides and downsides is great opportunity to reevaluate. Thanks.

  20. Deena says

    We cut off sleeves from extra surgical gowns and slip them over the AAH – they cover up the stick handle and your own hand and arm. It creates a much more natural picture, especially if the sleeve you use matches your clothing. Can also pick up men’s shirts from thrift stores for sleeves. Thanks Trisha for reminding everyone to follow a strict protocol and not harass dogs!

  21. LisaW says

    If only 3 out of 180 dogs in Diana’s research have snapped at the hand, but “the hand test has a high false positive rate and she suggests retesting for any dog that fails but does ‘well’ otherwise.” What constitutes a fail and how many have ‘failed’? Is there a scale of failure?

    Judging from my own experience and others’, I thought many (not all) shelter dogs had some issues or stressors. I guess I’m drawn to the problem ones :-)

  22. says

    I hate seeing dogs provoked with am AAH – so many people are way too pushy with his tool, knowing it won’t hurt them if the dog bites.

    There are a lot of good resources on on using an assess a hand properly, and on working with dogs who are uncomfortable sharing their stuff. Food guarding can be a pretty easy problem to work with, especially in its milder forms. Marder’s study suggests that about half of dogs deemed guarders in a shelter won’t show the behavior in a home at all. How many great dogs are missing a chance at a home because of overly enthusiastic evaluations?

  23. Tricia says

    Assessing dogs is only as valuable as the person doing the assessing. Trying to have people go down the list who don’t read dogs well does not work. If we place dogs in a family with kids that bites kids, it is only bad news for dogs of the future. The less we serve adopters, the more people will cease to go to shelters for dogs. We need to admit that there are dogs out there that don’t belong in an average household, or we are cheating people and dogs. The tool should not be the focus. The evaluator needs to be the focus. And there are lots of people evaluating dogs that don’t see dogs.

  24. Beth says

    Here is the comment I can’t post from home.

    Love the video clip and I may have to bookmark it to use the next time someone in an online dog community says their dog resource guards so he must be “dominant” and needs to be shown “who is the boss.” Ouch.

    Puppy Jack growled if you moved his dish when I brought him home, so I used a truncated version of the ASPCA’s recommendation and now he thinks people going by his dish must be a good thing. I got the impression he was growling at the BOWL (How dare you move away while I’m eating from you!!!) and not me, but best to be safe. Maddie eats faster (if that is at all possible) if you go near her. Mostly I leave them in peace to eat; occasionally I drop something in their bowl so they remember that it’s good to have a person come by their food.

    I can see where the hand would come in “handy” (pardon the bad pun) because some dogs are fast and either not inclined to warn by nature, or were punished for warning and so act without a warning. I can also see how it could be abused by someone with a mean streak or just a little clueless. It is important to know whether dogs are resource guarders before sending them home, I think. Some otherwise sweet mellow dogs can be very intense guarders.

  25. HFR says

    I just watched all the item videos on the ASPCA site. Really interesting stuff, especially because by the time you watch all the demo dogs you definitely feel as if you have a good idea of each dog’s personality, at least in a very initial way.

    The step that worried me more than the AAH was the “sensitivity” portion. Boy, those look like hard grasps of the dogs body, especially in the instruction to twist the skin at the same time. I’m surprised there is no mention that the dog may have some physical problems if they react badly to this item. It just looks like it could easily hurt, even more than the squeeze item.

    Otherwise, it’s clearly a very thought out and detailed assessment and has to offer a much better evaluation of a dog’s temperament than just trying to observe general behavior in a shelter setting.

    Thanks, once again, Trisha, for a fascinating blog post!

  26. says

    I don’t understand why there is a necessity for it to ‘look’ like a hand if the reality aspect is neither here nor there – particularly when we know that the primary sense is olfaction?? This article states:

    “I’m not convinced that it matters whether the dog perceives the hand as “real” or not, as long as the dog perceives the hand as something that he has to guard against”

    ….. In that case, why not use a rubber snake, a crutch , a twig or a piece of hose pipe?? What is the benefit of purchasing a cold pretend hand that smells of rubber and is inanimate? … Surely the money would be better spent on enrichment?

  27. Trisha says

    To Caroline: It took me awhile to figure out the “you have white rugs!” comment. I figured you must have been commenting on someone else, until I realized that the photo of Willie… who is on his sleeping pad, looks like I have a white rug. Ha! Not a chance! I have a bunch of soft pads for the dogs, that I put over the carpet (an ancient, thread bare oriental rug that I’ve wanted to replace for many years now) in the living room. Every once in a while I throw the pads in the washer, after combing out bushes of dog hair, a few burrs, etc, and they turn back to creamy white instead of farmhouse brown.

    To Tricia: I agree completely that, even with a clear test with consistent guidelines, the evaluator is the key. I suspect this is where a lot of shelters and rescues have problems. I’ve seen… we’ll, you know. The challenge is getting enough people well trained to do the work, especially given the limited time and budget of so many shelters. That brings me back to my ultimate goal: Less dogs go into shelters in the first place, shelters become like community health centers that focus on wellness and prevention. Someday…

  28. Kat says

    I’m puzzled by the black and white nature of the assessment, either the dog resource guards or the dog does not. This is not the day to day reality I see with my dogs. Some days Finna, especially but Ranger to some extent, resource guard and most days they do not. Most of the time anyone in the family can step over Finna when she has a bone and the cat can wander around past her and the only reaction is an ear flick but some days she’ll growl if you come near her when she has a bone. The Great Catsby and Ranger can both tell at a glance when Finna is in a resource guarding mood and will turn and walk away instead of entering the room so I rely on their assessment or wait until she tells me. Both dogs will give me anything they’ve found in the house that they want to chew but they might not want to give up their bone although mostly they will. I guess I’m wondering how accurate one or two data points on a single day really is as an assessment of overall behavior.

  29. liz says

    I can’t seem to carve out enough time to write a thorough comment and it’s frustrating. Nevertheless, thorough or not, I have some thoughts. In my neck of the woods, breed-specific rescues usually pull dogs at risk of euthanasia regardless of temperament. (Severe bite cases exempt.) These same rescues usually do their own evaluations, and develop their own treatment plans accordingly. Given the limited resources, and “the bad and the ugly” as described above, should overcrowded shelters be spending any time at all testing dogs who are likely to have a spot in breed rescue? Is a test unnecessarily stressful and redundant? Or is the evaluation a chance for socialization, a break from the kennel, and a look into a dog’s personality in another context?
    I think about stuffed stimulus dogs, as mentioned in my earlier comment, because I’ve seen it used as one stage of dog-dog assessment, and it gets a much different reaction from people than the AAH does. Onlookers are borderline enamored with the stuffed dogs. We name them, jokingly pet them, and I think we secretly want our own. Meanwhile, the associations we all have with a hand on a stick is nothing remotely similar. So although a life-sized stuffed animal could be even more frightening than a hand to some dogs, it’s treated as a benign object.
    An additional reason that commenting on this issue is tough is because it has no end. One thing leads to the next, and questions and problems remain hanging without resolution. And opinions differ greatly. I have at least four other topics I could wander in… All I can say is that I appreciate the focus here on the blog in that it offers a certain amount of clarity in a sea of questions and often disagreements. It also encourages more people to be seriously interested in looking at behavior objectively since that task is almost unnatural to us. I would have to say I’m skeptical of anyone’s absolute certainty of temp test results unless they’re just listing behaviors. But even among those who have been reading dogs fluently and for many years, we are human and they are canines! We have off days, dogs throw conflicting signals, and we are but limited individuals. Two women performing a test with an AAH would tell you nothing of what happens if a man was in the room (in some cases). Sigh, I’m wandering. Ultimately, an AAH is just a tool, and so is the whole behavior evaluation, which need to be regarded with utmost care.

  30. EmilySHS says

    I totally agree that the experience of the evaluator is key, and one thing I’m supremely grateful for is that at the shelter where I work, the formal eval is just one slice of the pie in getting to know the dog. I will say, though, that after a tester has done a whole bunch of tests it becomes a very interesting slice. Because the test items are standardized, testing becomes a “body of work” to compare responses across a large sample of dogs. Most responses are exactly what we’d expect–average, normal, in the fat of the bell curve. Outliers aren’t necessarily bad, but if a dog gives me something extreme or non-standard, it does suggest that it’s worth looking into a bit more.

    That said, I’m not sure the results have to be a matter of opinion. I deeply hope that anyone doing a lot of testing with a lot of dogs keeps track of their personal “batting average” in calling it correctly–keeps track, for example, of how many dogs they tested went into homes and snapped or bit someone, did poorly, did well, got returned, etc. My own feeling is that the test results are predictive of two important things that help me to make better placements: 1) how the dog is likely to behave when initially meeting strangers/potential adopters (shy, fearful, rowdy, touch sensitive, etc.) and 2) issues that may crop up immediately after adoption. Because really, besides the usual suspect triggers of having paws handled, being hugged, food bowl, etc., I think what I’m getting is the dog’s off-the-cuff response to novelty and stress. How the dog will behave in a month or a year may be anyone’s guess and highly subject to learning and environment, but to give the dogs a shot at getting into a new home, I at least need to know what to expect when they’re meeting new people in a stressful and novel situation.

  31. HFR says

    I was talking to a friend on the train this morning on the way to work. He has a Rat Terrier and a 4 year old child. I was telling him about this blog discussion (he loves dogs and has strong feelings about them, but is not that educated on the latest theories of behavior). Anyway, as I’m telling him about the controversy surrounding the use of the AAH, he takes off his glove to show me how his dog bit him a couple of times recently when he tried to take something away from him and didn’t have a treat handy to trade for it. He was very nonchalant about it. Kind of like “Oh yeah, he does that all the time. We just know to stay away from him when he growls.” I asked him about his child and he said she also knows to stay away from him when he has something he doesn’t want to give up. I asked him what about when she was really little and didn’t know, he said they just kept her away from him when he had a toy or a bone. He also said the dog and the child are quite close and often sleep snuggled together.

    Obviously, I was dumbstruck. But it just goes to show you how management can be more important than we think and how we can never know how a dog will do in any particular home. Clearly, we can only just try to do our best to keep any dog from doing any kind of harm to anyone else. We only have the tools and instincts we have. We can only try.

  32. Jacqueline says

    Speaking of the Stuffed Stimulus dog. What is considered a “bad” or “good” reaction to it. My Simon was at dog class at our local MSPCA and they had the stuffed dog out from a previous evaluation. Simon was freaked out by it the first time. And approached it another time it was left out. He barks, jumps back, goes forward, play bows, jumps backwards, barks, etc. I clicked and treated him when he approached it with no barking. He eventually touched it with his nose. But was still very wary of it. It was obviously not a real dog to him. It did not move or smell right. So, how do you expect a dog to act to such an artificial stimulus?

    He is also scared of the vacuum. He barks at that, then jumps forward onto it so it moves. He seems to be having a great time moving it, tail going a mile a minute. He pushes it across the whole room.

    And balloons, scared of them too. Anything new really. Then once he sees it is fine, he tries to play with it.

    He seems so conflicted. Would he fail the temperament test?

  33. Robin Jackson says


    My own concern when hearing about families with dogs that bite hard enough to leave marks who seem nonchalant about it is this. When a family has a child, they almost always have a lot of child visitors as well, whether they’re relatives or friends of their own child. Yes, you can deal with some dog/child issues through training your own child to keep their distance from the dog at certain times. But the visitor children won’t have any of that training. So I am uncomfortable relying on any management plan that depends on young children knowing the right thing to do. But to be fair, many of these families do put the dog in a separate area when visitor children come over. So it’s not impossible, but it does concern me.

  34. HFR says

    @Robin: I was very concerned and told him so. I actually told him I would try to find someone who could come to his home to work with him on this issue. I told him it’s fixable, which I suspect he didn’t realize. He was very open to that. That’s a good point about visiting children and he does put the dog in another room when they do have any visitors, but still, one slip up and there could be a serious situation.

  35. liz says

    Jacqueline, I wish I had more information about the Stuffed Stimulus Dogs.
    From my experience, “good” and “bad’ reactions are the extreme ones: overly friendly or aggressive behavior. In theory, the SSD is used as a first step in an assessment to prepare for how the next step, the actual dog meeting, might go. Although most dogs behave between this spectrum, prolonged agonistic displays and/or biting the SSD are the only things that make a dog fail this first step.
    Overall, my own generalization (for whatever it’s worth) is that dogs respond in a variety of ways to the posture/stimulus from a distance, but the majority show some hesitation/conflict/anxiety upon closer examination. This initial, from-a-distance reaction can be safely achieved with a real dog so …I’m not sure of any gain to anyone there. (Perhaps avoiding stress on the helper dog?) Many dogs who display fearful behaviors at any distance to the SSD are fine with real dogs. Many dogs, perhaps ones familiar with toys, react to being close to the SSD in a playful aroused state, like it’s the biggest (strangest?) toy they’ve ever seen. And this playfulness is not always representative of how they are with real dogs.
    I’d love to read anything more about the SSDs as I struggle to completely understand their usefulness in a behavior evaluation. I know that going into assessments I initially thought that the fake hand would be more of a “trigger” than a stuffed dog. But like your guy, many, many dogs don’t seem to know what to make of the thing. Yes, dog-dog assessments, like guarding assessments, are potentially hazardous and that a tool may help avert some of the danger… but I just need more information to be convinced it’s indeed helpful.

  36. Jacqueline says

    Thank you Liz. Simon is good with other dogs, but I need to always manage it. We had him one week (13 weeks old) and were walking him on a lead around our block. A neighbor’s beagle got out of their yard and attacked Simon. I spent a lot of time reinforcing that not all dogs are mean. He has been fine and happy will all dogs he has met at dog class. And many other dogs we have since met, there are still some he won’t approach so we quickly move on. He was fearful for awhile and I worried that it could turn into aggression. I always inform other dog owners that he may be fearful. Once he has touched noses he now goes into play mode. This is almost as bad as when he hid behind me meeting other dogs. Now I have a bouncy 9 month old 40lb dog being nutty.

    His reaction to the stuffed dog was not how he reacts when nervous about a new dog.

  37. HFR says

    Just FYI, the ASPCA Safer test does not include the stuffed dog or baby doll tests. At least not that I saw…

  38. Ingrid Bock says

    I will read this again later, at which point I will probably mightily regret the following comment, but maybe not–maybe I will have the courage of my convictions. (What’s that quote? Something like, ‘It is a poor thing, but it is mine own.’ I suppose it’s good to admit to one’s truths.) There are five books which form the cornerstones (maybe the fifth one could go up at the top) of the building which stores my core values and beliefs about how we should interact with dogs. I was fortunate in that they were also some of the first books which were recommended to me, or chosen by me, as I began my organized efforts to study dogs. The five books are, ‘The Other End of the Leash’, ‘The Culture Clash’, ‘Don’t Shoot the Dog’, ‘Plenty In Life Is Free’, and, ‘If A Dog’s Prayers Were Answered, Bones Would Rain From The Sky’. You mention the AAH and Sue Sternberg in ‘Other End’, I believe. It struck a false note for me then, and it strikes even falser now that I know more about shelter assessments and their consequences. Assessing dogs in shelters for resource guarding does more harm than good. The bit about the AAH and Sternberg is such a tiny mention in your book, if I’m remembering correctly–just a sentence or two. And yet it glued itself to my brain and heart, and prevented me from fully trusting the book. There. That’s what I may mightily regret later. I’m afraid to hit, ‘Post comment’, but I will. Your description of the assessment of the tied-up dog here fills me with rage and sorrow. These things should not be happening. You are in a very powerful position to educate. I will be cheering you as loudly as I can, should you decide to use your influence to work hard for shelter assessment reform.

  39. Trisha says

    To Ingrid: Thank you for your thoughtful comment. I very much appreciate your honesty. Not always easy to do and I give you credit for it, especially about issues that create a lot of passion. First off, I’m glad that we are on the same page regarding improving behavioral assessments of dogs. That is exactly why I wrote this blog, because it is important for people to understand that the tests can be done in such a way as to 1) result in inaccurate evaluations and 2) create problems where none existed before.

    My answer to your expressed concern is this: I am someone who has never seen the world in black and white. Thus, I have great respect for many people who I don’t always agree with. Sue Sternberg is an example of that. I don’t always agree with everything she says, just as I don’t agree with everything just about anyone ever says. Including myself for that matter: Sometimes I look back at some of the things I wrote and either shake my head or want to bury it. I know Sue, and I know that she loves dogs deeply, and has dedicated her life to improving theirs. I don’t always agree with her evaluations, but I also know that she has done tremendous good in the world. Remember that when she started doing assessments, just about every shelter “euthanized” large numbers of dogs, and did so randomly. The sweet, adorable plain brown dog was put down because it wasn’t as cute as the dog that was clearly aggressive. There was simply no attention paid to anything but looks, or how long the dog was at the shelter. Since that time, massive changes, many for the good, have occurred. Fewer dogs are killed, more are placed, many shelters have well-trained behavioral staffs that identify potential problems and work to turn them around. You may not like behavioral assessments, but I wish I could take you back in time to see what the world was like 20 years ago at shelters. I’m not sure you would recognize it.

    That said, if my mention of a name of someone you don’t agree with is enough to put you off everything else that I’ve said, even if you agree with it, there’s little I can do to change that. I do agree with you in one thing: Having the courage of one’s convictions is important. I’ve dedicated my own life to trying to improve the relationship between people and dogs, and sometimes it hasn’t been easy. I will continue to work toward a world in which both species are benevolent, wise and kind to each other, even though the best way to do that is not always clear. Thanks again for taking the time to write.

  40. LisaW says

    Trisha, you are grace incarnate. Your striving to see a “world in which both species are benevolent, wise and kind to each other” is evident in how you welcome such a wide array of points of view and respond to our comments.

    I recently came across a blog on PTSD in dogs, and in looking at the author’s other blog postings, I noticed a link titled “My Letter to Patricia McConnell.” Of course I clicked on it immediately, and without dredging up the old conversation, your review and responses to the author’s opinions on how long to wait before “training” your dog were replete with wisdom, patience, and grace.

    I don’t know how you do it, and I know I don’t have nearly enough of it, and I certainly appreciate it.

  41. Trisha says

    To LisaW: How kind you are! This is why these conversations are so good, yes? To get the full picture. Thanks again for your generous comment.

  42. Karen says

    Thank you for this post and your ongoing efforts to educate shelter staff. We adopted a dog from a shelter that does NO behavioral testing (other than dog-on-dog reactivity), and it has made our life much the harder. Our sweet, smart and very scared dog has taught us much about dog behavior and she has absolutely made progress (no more fear of leaves, crickets, storm drain drains, vacuum cleaners etc etc). However, she remains scared and becomes aggressive with some dogs and some people.

    Despite intensive training, working with trainers, and medication, she is still not completely reliable around others. As a young family with a toddler and twins on the way, adopting a dog who requires so much care and attention was unexpected and it’s hard! We both grew up with dogs and were prepared for lots of walking, training and the basics. But this is a whole other ball game. The time, the cost and the worry about keeping our family, the neighborhood and other animals safe while providing a good life for our dog is hard!

    It’s not bad enough to put her down (although during her most volatile days we considered it); however, finding someone who is able and willing to take a dog like her is also elusive. I know it’s often mentioned as an option, but we have never found anyone who would be willing to do so (and she really is sweet, loving, smart and adorable when she is not scared!). So, we continue to make the best of it and to balance appropriately all the needs of our family. However, I wish, I truly wish that we would have had all the information prior to the adoption and been led to a dog that was a good fit for us at this moment in time. It would have been better for all involved.

    Please keep up the good work. I hope it helps other dogs and families!

  43. Debby Rightmyer says

    Until recently I never realized that the assess-a-hand and evaluating for RGing was controversial. How in the world can you appropriately place a dog without some feedback on this??? I have seen direct correlations with dogs that show RGing during an assessment and in real life scenarios (based on bite history’s, surrender questionnaires, general observations, etc.).
    Of course it is not an exact science, but you have to make an effort to determine likely behavior. And like Patricia said, it isn’t black and white. While many dogs show some level of RGing, they may be placed with some limits on age of children and the like. Other than shelters with high euth rates, I don’t know any shelters that just euth Rgers (and those shelters are also euth for uri to try to make stats look like no kill), most use that to determine the appropriate household. Some may euth Rgers depending on level of Rging, predictability, generalization, warning signs shown, and if the dog leaves the item to bite.

    It appalls me to hear that many are using the tool and assessments in general to push dogs to react.

    To dismiss the test because it isn’t 100% is like saying cars shouldn’t be safety tested because different situations can cause different outcomes and there is no guarantee.

    I too can learn from somebody that I don’t agree 100% with. Very few people do I agree with 99-100% of the time, Patricia you are one of them, I mean that sincerely, I truly do value your opinion and think you are one of the most sensible people I know.

  44. Trisha says

    Thanks Debby, so gracious of you. I’m not sure that everyone who knows me would agree that I’m sensible, but I try!

  45. Jacqueline says

    I think any information is good information when adopting a pet. I was just not sure of the validity of such an artificial device. I know my dog would be freaked out by the hand, although we can put our own hands in his dish with no issues. If it is just used to gather information that is fine. But, if the dog is scared of IT and it is not Resource Guarding, that needs to be determined.

    We adopted a cat from our local spca once. We did not get full disclosure on this cat. She was growth restricted, and had many issues. Our vet thought she was brain damaged and basically stuck as a kitten. She could not figure out how retract her claws once out and never figured out how to use the kitty litter, among other things. We ended up giving her a good, but restricted home. We knew if we returned her to the spca she would never be adopted. We never really bonded with her like our other cats. I would have prefered to have all the information on her before hand as well. It really turned us off adoption since many of her issues should have been noticed, even in a shelter environment.

  46. liz says

    First, my apologies for any hurriedness to my tone earlier. Second, I agree with all comments appreciative of Trisha’s work and approach and graciousness, and I’m so grateful for that influence in the world.

    The controversy surrounding behavior evaluations (and their tools) can change depending on the context, and so can the intensity of the controversy. There are evals for proper placement, evals for euthanasia, and times where both settings combine: if a dog scores poorly in too many areas and the adopter pool becomes too limited, then time/space/behavior all mesh together and dogs are killed. Controversy abounds in the predictive value of tests, whether or not appropriate and/or effective methods are used, and finally, in the consequences of the test results.
    I think that the biggest controversy, as one might imagine, is with evals that determine whether a dog is at all placable. I also think that ultimately many of the lower tiered controversies come back to the life or death consequence, as showed early on in this post’s comments. If the consequence is a relatively happy one (e.g. if a dog guards in a shelter, is carefully placed, then goes on to show no guarding tendencies) then the surrounding controversy of the false positive can become an argument against instances where the outcome is death.
    Regarding the predictive value being less than perfect, controversy also diverges into one of adopter satisfaction/disclosure. I keep thinking of the article in the most recent BarK magazine about veterinarians (if I remember, there were a couple articles actually, my apologies for forgetting the author) where the first page of the article describes what we hope for in our veterinarians. The author describes our entirely reasonable desire for usually unattainable super-human qualities. I think the same is true of animal shelters, and this notion of everyone rightfully seeking shelters full of superheroes resonates with me. Along these lines, I continue to think of the recent controversy surrounding eliminating dog-dog intros with resident dogs, and that one of the pitfalls besides lack of predictive quality is that the meeting sets false expectations for adopters. If you apply the same thinking to the behavior evaluation as a whole, then how can shelters more effectively address adopter expectations?
    I don’t think temperament tests should go away. But I hope that someday we can find a suitable, improved alternative, and find a way to be more superhero-like in the mean time. Another of the lower tiered controversies, at least in the non-confrontational training community is that the tests are designed to be stressful. (There are a variety of test protocols, and some use multiple tests, others use a smorgasbord of what are perceived as the “best” elements of several tests.) This induced stress is counter to so many principles of training, yet is currently unavoidable with the system as it stands. This paradox is reconciled in a variety of ways, but I do fear those justifications fuel the fire of those pushing dogs to elicit reactions. So if a dog then fails after being pushed and is euthanized, then this lower tiered controversy of induced stress jumps up to the realm of biggest controversy after all… I also fear that these justifications of necessary induced stress are one of many things which can affect an evaluator’s ability to score in a neutral manner day in and day out. (Which, btw, is another superhero expectation- where testing is to be serious and scientific yet nonchalant and neutral when there is so much weight riding on the situation.)
    There is just so much room for error and expectation problems that I hope a better solution is around the corner somewhere.
    It is quite possible that I’ve spent too much time trying to help improve an open admission urban shelter with approx. a fifty percent kill rate, and that my perspective is altogether jaded, but my hope is to clarify a few things and think things through… fool’s errand;)?!?

  47. AAB says

    Thanks to everyone for such a thoughtful discussion!

    Those who raise questions about the validity of evaluating a dog’s “aggressive” tendencies when faced with an AAS, a novel (and, when it comes right down to it, rather bizarre) stimulus are making an excellent point.

    However, I believe that this demonstrates one of the truly appropriate uses of an AAS: how does the dog react when faced with new and strange situations?

    I’m lucky enough to work in a shelter where we’re able to work with “aggressive” dogs for as long as it takes to address the fear and stress that leads to the aggression in question. We have the resources to reevaluate the dog periodically to determine its progress under its mod plan. I’ve seen dogs (and cats too) make amazing turnarounds because of the work we do.

    We absolutely use an AAS for the intake evaluation, but we *never* label a dog as “aggressive” based on its reaction to this object. We note that the dog may react fearfully to new objects, particularly in a new and stressful environment, such as a new home.

    This is where the value of the AAS comes in. If a client wants to adopt a Fido, I can look up his file, and see his reaction to the AAS. Let’s say he snapped at it – an understandable reaction when one finds oneself in an animal shelter, surrounded by weird smells and sights and sounds, having a strange object reaching into his personal space.

    Now, maybe Fido does much better on subsequent evaluations. I can let the client know that while his behavior improved dramatically during his stay, Fido may be stressed and snappy when he’s faced with the new home environment. I can advise them on how to reduce his stress levels, easing his transition to his new home.

    But let’s say that Fido continues to be react to the AAS. I can now inform the client that Fido may need confidence-building when having his personal space invaded. This is very valuable information that I can use to counsel them on how to address this issue.

    The client then has the knowledge they need to decide whether they can commit to the time and effort it may take to desensitize their dog, or whether to go home with one that’s more bomb-proof. Should they decide to take on the dog’s challenges, I can send them home with the appropriate resources to work with their dog.

    I wholeheartedly agree that a dog should never get the Scarlet Letter (A for “aggressive”!) just because of its initial reaction to the AAS. It is, however, a valuable tool that provides information about the dog’s behavior; and, as all of us know, the more information we have, the better we’ll be at modifying the dog’s behavior.

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>