Physical versus Psychiatric Service Dog Training

Service Dogs can be trained to help mitigate many types of disabilities.  Some dogs are taught to perform a physical activity like pressing a button to open a door, providing mobility support or retrieving a fallen object.  In that sense, the physical service dog replaces the assistance that another person or a device might deliver.   A grabber stick can often be as efficient as a dog at fetching an item and a cane requires far less space and cost than a mobility support dog.  Still, there’s no way to artificially replace the companionship a dog provides.  That relationship can be even more important for a person’s health and well-being than having a means of opening a door.  There are benefits which reach beyond the physical tasks that a Service Dog performs, which can be considered highly valuable freebees!

Service dogs can also be utilized by people with psychiatric disabilities.  These individuals may not require physical assistance from their dog.  However, that doesn’t mean their lives are not significantly enhanced through their partnership with a trained dog.  Currently, the ADA does not make a distinction between dogs which assist with physical tasks and those which are trained to perform tasks that mitigate aspects of a psychiatric disability.   Their handlers are granted equal access rights when traveling in public places.

In my personal experience and opinion, unlike many physical service dog tasks, the assistance that a psychiatric service dog provides his handler cannot be provided by an alternate source.  Even a highly devoted individual like a parent who loves their child more deeply than life itself, can be perceived as being “put out” when assisting their child.  That goes for even very young children who do not want to feel that they are a burden. It can add a level of anxiety above and beyond the actual disability.  A dog doesn’t have a job, doesn’t need to tend to sibling or make dinner for the family.  A dog has one desire and that is to partner with its handler.  A dog can alleviate the angst of requiring assistance from another person. Additionally, mitigating some psychiatric conditions cannot be accomplished through physical assistance. There’s a bit of “magic” that happens when a dog serves in the capacity of a psychiatric service animal.

There are many tasks that a psychiatric service dog can perform.  Some individuals are assisted when the dog does physical tasks such as laying across her lap or feet to offer a type of deep pressure therapy.  Dogs can also be taught to remind a person to take meds (preferably as a back-up to an electronic alarm.)  In various ways, like being trained to accept grooming or laying its head over the handler’s lap, the dog can calm a person who is having an anxiety attack.  These, too are physical tasks.

In my experience, a highly beneficial task that many psychiatric Service Dogs perform is mitigating what I like to call a “shift-from-normal” event.  I define that as anything from a seizure, a PTSD episode, a panic attack or the onset of behavior which causes self-harm.    Determining the sorts of tasks a dog can be trained to perform in order to mitigate a shift-from-normal event brings forth challenges which are fairly uncommon when training a dog to perform a physical task.  

As a trainer I don’t have possession of the trigger for a shift-from-normal event.  Therefore, I am unable to make a direct connection between that cue and the behavior I want the dog to perform.  I  don’t have a jar of “panic induced anxiety scent” or a boxful of “PTSD episode dust” sitting on a shelf ready to present to the dog.  For that reason, I  have to back my way into teaching a dog what I expect.   I am aware, through observation and reports from dog owners that dogs can perceive shifts-from-normal events.  Perhaps the most notable are seizures (epileptic and otherwise.)  There are reports that dogs can alert to a seizure event up to 30 minutes or more before its owner presents with the episode. 

It’s my belief that if some dogs can detect the onset of a seizure probably most dogs can do so.  Unfortunately, we don’t know what or how they are sensing.  Unless we can come up with a method of communicating that we want a dog to tell us when he perceives the upcoming event, most dogs don’t alert us, even if they can.  Also, even if most dogs can sense the impending episode, I suspect that not all dogs care enough about their owners to do so.  Some dogs may believe that watching squirrels or hunting rabbits is higher on their daily to-do list than communicating with their owner who may lack in canine leadership skills – so their dog just “checks out.”  This is why I believe selecting the right dog and training it to very high standards for service dog work is so important.

When we don’t have an actual cue (aka “panic attack powder”), we can follow these steps:

1. Teach the dog a response behavior, like nudge me with your nose.
2. Achieve the level of performance where treats have been significantly reduced or eliminated, and the dog performs the behavior even in the presence of distractions.
3. Any time the handler senses the onset of the shift-from-normal, ask the dog to perform the response behavior and profusely reward him. 

In the best possible scenario, the dog quickly makes the connection between the shift-from-normal event (that it naturally senses) and performing the nudge task.  This is expedited when the handler presents with fairly frequent and consistent episodes.  That allows the dog to make the connection more easily.  Hoping to receive his high value reward sooner than later, the dog will begin to alert (as in prior to the human-perceived onset) rather than being asked to perform the response (once the episode has begun.)

Some pitfalls which happen when we don’t have the actual cue to present to the dog are:

1. The handler doesn’t present with shift-from-normal events frequently enough for the dog to engage in practice sessions.  It’s great for the handler if her seizures are sufficiently under control that she only experiences them once every six months.  But, that may not be enough training time for the dog to understand that we want him to perform a task at that time, and it’s probably impossible for the dog to learn to alert in advance of the event.
2. The dog learns that the shift-from-normal is significant but chooses to present a wholly different behavior than the one we taught him.  For example, a dog that has been trained to nudge his handler might rather woof, spin in a circle, scratch at the carpet or jump up on his handler.  That behavior can be perceived as unacceptable or even worthy of a correction.  Sometimes, the dog is sent outside because he won’t stop circling (may even appear frantic) and then moments later his handler presents with the shift-from-normal event.  It’s not uncommon for the people in that situation to fail to make the connection that the dog was alerting to the impending episode.

Clearly, having a dog alert to a shift-from-normal, even if it isn’t with the trained behavior, would likely be considered a win.  But, since the dog wasn’t “trained” to spin to alert to an imminent panic attack, does the dog meet the ADA’s requirement that tasks must be trained?  That is a topic for another day.

The take home message to this post is that training a Service Dog to perform a physical task is not the only way to produce a dog that will help mitigate his handler’s disability.  Even if we don’t fully understand how dogs can sense a looming event, there is plenty of evidence that dogs can and do respond and alert to what I refer to as shift-from-normal episodes.  We humans are able to sense energy shifts in other humans (small children know when a parent is “off” even if they don’t know why and the parents tries to hind their feelings from the child.  As adults, we usually are taking too much to listen to energy shifts! Dogs are very good at it. That they are willing to share it with us is an amazing gift.