Monsters in the Closet

The development of superstitious associations.

While we don't usually like to compare people to dogs- we are, in fact, different species- we do share similar psychological processes.  To help you understand superstitious associations, I want to put everyone in the dog's shoes in this example because the psychological process that the dog undergoes would be the same for a person.

Skippy's Tale

The Scene: 

On a quiet cul-de-sac in Suburbia, USA, a dog, Skippy, rests contently under a shaded tree.  A bird chirps, a breeze blows; the dog closes its eyes and lifts its nose to take in the wind, his new electric fence collar tightly secured around his neck.  Skippy's owners like to let him outside but don't want to put up an unsightly fence- this new electric fence technology is perfect!


Act 1: 

As Skippy observes his environment, he notices a strange person walking past the house.  As this new person approaches, Skippy feels the urge to investigate- he doesn't usually mind a human-friend taking a nice stroll past his house, but he's never seen this person before.  Could be danger, could be food, could be nothing, but he doesn't know until he checks it out.

As the person approaches, Skippy approaches the property edge.  


Act 2:

Skippy is interested in this potential new human-friend and attempts to approach them in order to investigate further.  

With one extra step... *bzzt* - "YIP!" and Skippy scurries back under the tree.

Skippy does not understand why this human would hurt him like that- he just wanted to meet them.  With a watchful eye, Skippy tries to act unassuming and lies still until this mean new stranger leaves the area.


Act 3:

Over the course of the next few weeks and months, Skippy becomes more wary of new-comers to his previously quiet cul-de-sac.  A new family moved in next door, and their child likes to bounce a scary basketball around too much in the driveway.  

Multiple times, Skippy has tried to keep these new people away from his yard- the last new person hurt him and he won't let that happen again.  Sometimes Skippy also gets hurt by these people as he runs up to the property's edge barking at them to stay away.  He can't understand why these people keep hurting him.

One day, as the child is playing with the scary basketball, it rolls closer to Skippy's yard.  This is it- Skippy has had enough.  He yells at the child "don't come any closer to me!" but the child doesn't listen.  Skippy gets closer, barking, but the child is not heeding.

As Skippy approaches, the child continues running toward the ball.  Skippy feels he is in danger and does the only thing he can think to do- barking didn't work anymore and he NEEDS this scary human to get out of his space.

As the two meet at the ball sitting just outside the property line, Skippy bites the child in the face.

These stories usually end tragically for all parties involved- the child is sent to the hospital, the dog to the vet, and only one comes back alive (hint: it's not the dog). 

This is the dog that would be labeled "unstable", "dangerous", "a loose cannon", and "unpredictable".  The bite presumably "came out of nowhere" and was "out of the blue".  Because of this, the dog can never be trusted again and is euthanized. 

But did it really come out of nowhere, or were there precipitating reasons?


A superstitious association is an inappropriate attribution between a stimulus and its perceived effect- the incorrect development of a "cause and effect" relationship.  

Superstitious associations are most often derived out of ignorance, anxiety, and fear.  Something (A) happened and it must be because of Other Something (B).  Even if B did not cause A, the person or dog's psychological state or ignorance perceived that to be the case.  Now A is inappropriately associated with B. 

The reasoning?

In particular, the inability of individuals—human or otherwise—to assign causal probabilities to all sets of events that occur around them will often force them to lump causal associations with non-causal ones. From here, the evolutionary rationale for superstition is clear: natural selection will favour strategies that make many incorrect causal associations in order to establish those that are essential for survival and reproduction.
— The evolution of superstitious and superstition-like behaviour - Kevin R Foster, Hanna Kokko - Published 7 January 2009. DOI: 10.1098/rspb.2008.0981

When the potential outcome for an event is life-or-death, it is more likely that unrelated causes of that event will become associated with it.  

So let's break down "Skippy's Tale" and see if we can pinpoint the superstitious association that caused an otherwise friendly dog to bite an unthreatening child in the face.

In Act 1- we know two things. 

  1. Skippy is wearing a new-to-him electric fence collar.  We can presume, since it is new, that Skippy has not been appropriately acclimated to it- his owners have not taught him how to avoid the shock. 
  2. Skippy is focused on something particular- the new person.

Superstitious associations are most likely to occur when the dog is focused intently on something in particular.  On top of this, his owners have not properly trained him how to avoid the punishment he would receive if he approaches the property boundary.  Skippy is not only unfamiliar with the tool itself (what the shock is, how it works, etc.), he also has no idea how to avoid it. 

A quick segue into how escape/avoidance behaviors are developed:

Shuttlebox experiments were instrumental in our understanding of how escape/avoidance behaviors are developed.  This video shows how these experiments were performed.  A tone sounds, a shock is applied, and the rat escapes both by shuttling to the other side.  Soon, the tone itself produces the same "shuttle" response as the shock did.  The rat learns to escape the tone and avoid the shock.

Now that the tone produced strong fear, any response that terminated this conditioned stimulus would be strengthened through the process of negative reinforcement. What response terminates the CS? Shuttling to the other compartment! Thus, according to Mowrer’s account, the subjects shuttle during the tone and before the shock begins, not to avoid the shock, but rather to escape the fear-inducing tone CS.

Note that Mowrer’s account appeals to two processes or “factors.” First, there is classical conditioning of fear to the tone, transforming it into an elicitor of strong fear. Second, there is instrumental conditioning of shuttling through negative reinforcement, the termination of the aversive and fear-evoking CS. Because it appeals to both classical and instrumental conditioning, the account is today known as Mowrer’s two-factor theory of avoidance.
— Instrumental Conditioning Avoidance Conditioning -

Now that we understand how escape/avoidance training works, let's figure out why it did not work correctly in Skippy's case.

How an electric fence is supposed to work:

  1. The dog is desensitized to the collar- this means it wears it all the time (supervised) with it off with no consequence for a couple weeks. It's just another collar that feels funny but is otherwise inconsequential.  The end goal is the dog associating the conditioned stimulus with the shock, not the collar with the shock.  Dogs that recognize that the collar delivers the shock are what are called "collar-wise".
  2. Flags are placed strategically around the perimeter of the yard, forming a visual barrier between the "containment area" and the "exterior area". 
  3. The dog is taken to the boundary edge, this time with the collar activated.  As the dog approaches the boundary, a tone sounds, warning the dog of the impending shock (though it doesn't know that yet).
  4. Most importantly, as the dog is walked toward the boundary, it is in direct line-of-sight with one of marker flag.  This is extremely important because we need the dog to associate the tone and shock with its proximity to the flag.
  5. The dog is walked forward such that is exceeds the "tone" radius and enters the "shock" radius. "OUCH" - that flag just bit the dog.  The dog MUST be able to escape this shock by retreating back into the yard.
  6. Just like in the shuttlebox experiments, the dog will learn to escape the tone to avoid the shock.  
  7. THE FLAGS and the dog's conditioning of the fence helps to prevent superstitious associations, though it does not completely eliminate the chance that these associations will occur later on.  

So, to break it down:

Tone + Proximity to Flags = Shock  

--> Tone = Avoidance of Flags

----> Dog is contained in yard

Many people that install and utilize electric fences do not appropriately condition the dog properly.  They assume that by placing the flags and allowing the dog to subsequently get shocked that the dog is appropriately associating the shock to the flags.  Some dogs do learn this but only by chance, and many dogs also form superstitious associations when the perceived causative stimulus is coupled with anything other than the flags. 

People also then tend to remove the flags.  The flags should NEVER be removed.  The flags are the only true indicator of the potential shock- without them, the dog no longer has a reference point.  It will then do its best to not get shocked, possibly by associating the shock with a certain tree or rock in the yard as its new reference point.

This is not fair to the dog and can breed extreme levels of anxiety and potentially dangerous behaviors.  We WANT the dog to understand the meaning of flags.  We do NOT want the dog to also be wary of the natural world.  Additionally, in the relearning process of where the 'boundary' is, you are at a high risk of potentially dangerous superstitious associations, such as other people and dogs.

In Skippy's case, his owners did not condition him to associate the flags and boundary as the shock point.  Notice that flags were not even mentioned.  Why?  

Because Skippy's owner simply put this new collar on their dog (who otherwise generally stayed in the vicinity of the house, the collar was "just for good measure").  Instead, while the shocks created a boundary (retreating back into the "containment area" allowed the dog to escape the shock), the dog did not learn the discrete location of the boundary.  To make matters worse, one of his first instances of feeling that shock was directly associated with a person. 

Skippy formed a superstitious association between the shock and the person.  This created a negative conditioned emotional response towards people.  Skippy learned that barking sometimes made people go away (because now he is scared of them and wants to create distance).  But, because Skippy doesn't know where the boundary is, sometimes he gets shocked when he approaches the boundary to push away his stressor.

At the point that he bit the child, Skippy was...

  1. anxious about people,
  2. anxious about the ball being played with,
  3. anxious about people approaching his yard,
  4. and in fight-or-flight when the stressor did not leave the vicinity at his earnest request.

Skippy chose "fight".  The bite was in no way "out of the blue" or "unexpected".

Unfortunately, bites like this happen all the time due to inappropriate use of electric fences.

So why did Skippy go through the fence to confront the child?

Remember that an aversive can only be as unpleasant as the dog perceives it.  There are ways that your body can block pain, such as an adrenaline rush.  This is your, and your dog's, bodies' way of getting you through a potentially life-threatening situation: you can live with a severe laceration, but you can't live if you get eaten by the bear that gave you the laceration- adrenaline keeps your focus on fighting and fleeing and your body and mind can worry about your injuries later.

Dogs with anxieties are pre-adrenalized- they are ready for action and are close to or are in fight-or-flight in anticipation of their stressor.  

Coupling Skippy's anxiety with the close proximity of the stressor, the shock was simply a nagging blip.  Adrenalized dogs can and will ignore something mildly unpleasant to escape or subdue their stressor. 

Let's further apply our knowledge of escape/avoidance training and superstitious associations:

1. Scat Mats

Many times, people will apply aversives in areas where their dogs are frequently being "inappropriate", such as jumping on counters or furniture.  They use things such as scat mats or sometimes even mouse traps to "bite" the dog when it puts its feet on something inappropriate. 

Why is this unfair?  The dog does not know how to avoid the aversive, so the only result of the situation is punishment.  Counters are now "always bad".  Furniture is now "always bad".  For the average dog this may be nothing, but if you plan on doing any type of sport with your dog, creating a fear of putting their feet on something will be extremely detrimental to any of your training- remember that fearful responses are more easily generalized.

Dogs don't know that there are "rules" unless they are taught them- no, they do not instinctively know that they should not jump on the counter or ask permission to be on furniture.  Teaching a dog behaviors such as "off" (get your feet off of whatever they're on), "leave it" (don't eat the food in front of your face), and "up" (you have permission to be on this surface) give you a means by which to redirect your dog's behavior onto something more appropriate without being unfair to the dog in the learning process.

2. Bark Collars

While unreliant on superstitious association to work properly, bark collars frequently have unintended consequences caused by superstitious association.

This is frequently seen if the dog is crated with the collar on or if the dog is housed with other dogs with the collar on.  The collar responds to certain pitches and tones.  These pitches and tones can be replicated by crate noises.  If a dog is shocked for an unanticipated reason (we assume the dog knows that barking = shock), then it is likely that whatever the dog was focused on at the time will be related to the shock.  This is a superstitious association.  

If a bark collar is used on a dog that is housed with other dogs, it is almost certainly going to get inappropriately shocked.  This is due to the fact that bark collars are nondiscriminatory- they don't care if it was the dog next to the collared dog that barked, it will shock the collared dog.  Again, the dog was shocked for an unanticipated reason and is much more likely to develop a superstitious association. 

This becomes extremely dangerous when that related stimulus is a person or dog.

3. Leash Corrections

The power of the leash correction comes from the handler's ability to communicate effectively with the dog.  The leash is an extension of the handler- it's a physical way to say "hey you, not that, this".  Inappropriate or poorly timed leash corrections can create superstitious associations.

The most frequent use of a leash correction is when the dog is interested in something and begins to pull on the leash.

What the HANDLER thinks they are correcting the dog for:

Pulling on leash.

What the DOG thinks they are being corrected for:

Who knows!?

Why does the dog not know what they are being corrected for?  Because most times, the correction was not preceded by some type of compliance command.  If a dog is "corrected" for inappropriate behavior without being 'warned' (given the chance to avoid the correction), it is HIGHLY LIKELY the dog will develop a superstitious association.  Why?  Because it received an unpleasant feeling and did know where it came from or why, so that dog that it was interested in MUST have been the culprit. 

We create aggression out of curiosity or reactivity by inappropriately correcting dogs for behaviors they don't know are wrong and are without the knowledge to avoid the correction.  Not only does a leash correction adrenalize the dog, if the dog found it aversive, it will likely form a superstitious association between whatever it was looking at and the sensation.  This develops into anxiety which can turn into aggression given the right opportunity.  "But he was always fine with dogs before, I don't know what happened."  Inappropriate use of leash corrections is what happened.

Again, dogs should only be punished for disobedience of a known behavior- this means that the behavior must be discrete, proofed, and generalized before it can be corrected for.  If your dog can't walk down an empty sidewalk without pulling on leash, that is NOT the time to start correcting it.  

Teaching a dog to avoid a correction removes conflict and prevents superstitious associations.  Teaching a dog a true "heel position", not just your general vicinity, allows you to correct your dog for being out of a discrete position.  Teaching your dog a "focus" command, where the dog must take its attention off of something else and onto you, allows you to correct the dog for not complying with the prompted behavior.  Teaching your dog a "leave it" command, again, gives you grounds for correction.  All of these things are behaviors that have a "yes" or "no" answer- yes you comply = positive reinforcement, no you don't comply = positive punishment.  The dog has been taught leash pressure, understands how to escape the pressure, and understands how to avoid the aversive (by complying with the command).

This is not to say every dog needs leash corrections- this is to say that if you are going to use a leash to correct your dog, you MUST be aware of its pitfalls if your timing is off or if your dog does not understand why or how the correction happened.

4. Shake Cans

Shake cans are used similarly to scat mats in that they are a purposeful superstitious association.  

Dog sticks its head in the trash can = extremely noisy terrifying rattling can falls from the sky.  Now trash can = the sky is falling (the dog's proximity to the trash can made the sky fall). 

Again, while these tools can be effective in certain instances, this is probably the MOST likely tool to create an inappropriate superstitious association.  Why?  Because you have to throw the can near the dog in order for the tool to work.  What happens when, as the can is in the air, the dog takes its head out of the trash can and turns to see its reflection in the sliding glass door?  The can is already on its way- you've now set your dog up to be deathly afraid of its own reflection. 

This is the ultimate "monster in the closet". 

What's the bottom line here?

1. Electric fences can be extremely dangerous and, more often than not, ineffective when they really need to work.  Adrenalized dogs will run through the fence to get to whatever they are focused on and are unable to re-enter the "containment area" due to getting shocked coming back in.  Additionally, they are producers of superstitious associations and can easily breed anxiety and aggression. 

2. We must treat our dogs fairly and teach them more than train them.  Being taught how to live life to reap its benefits is more rewarding than constantly struggling to stay within bounds.  

3. If escape/avoidance techniques are to be used, the handler must have a firm understanding of the consequences.  Behaviors learned through escape/avoidance persist longer due to adaptive learning.  Mistakes made through escape/avoidance take much longer to undo and may be irreversible.  

4. When in doubt, assume the dog does not understand that it is doing something wrong.  More often than not, most dogs have been poorly generalized and they do not have a strong understanding of how to perform the prompted behavior in a given situation.  Removing the dog from the situation is more effective than inappropriately applying a punishment and prevents the possibility of superstitious associations.

Open the door and shine the light into your dog's closets and show them there's nothing to be afraid of.  

Promote interest and requests for learning.

 Treat them fairly and respect them as sentient beings that are a different species from us and do not inherently understand our world.  

Your relationship with your dog will be better for it!

Questions or Comments? 

Let us know!