SP6B NUISANCE BARKING IN DOGS AND THE WELFARE IMPLICATIONS OF DIFFERENT TREATMENT MODALITIES

Transcription

1 SP6B NUISANCE BARKING IN DOGS AND THE WELFARE IMPLICATIONS OF DIFFERENT TREATMENT MODALITIES Sara L. Bennett, DVM, MS, DACVB VCA Behavior Service VCA Berwyn Animal Hospital Berwyn, IL, USA Overview of the Issue Nuisance barking by pet dogs is a common complaint in urban and suburban areas. Many tools intended to stop this unwanted behavior are commercially available. These tools are advertised as effective to the general pet owning population, but often their use, particularly as a response to anxiety or distress related barking, seriously compromises the dog s welfare. What is Nuisance Barking? Cross defined it as barking that is of sufficient frequency to cause distress or interruption of the life of the dogs owners or other residents within the neighbourhood. 1 The problem is actually more with the people around the dog being intolerant of the level of barking displayed than with the inherent reason the dog is barking. Barking, which may be a normal behavior for the dog, is in many settings, socially unacceptable. Excessive barking can lead to relationship damage between neighbors, damage to the human animal bond and result in monetary fines, eviction, pet relinquishment, abandonment and euthanasia. Motivations for Barking- 1,2 There are many different reasons a dog could bark. It is a form of communication, both intraand inter-specific. Dogs can bark as a clinical sign of anxiety or as a clinical sign of other behavior problem(s). When performed outside of a normal context, barking is really a nonspecific sign of a behavior problem, similar to the clinical signs of vomiting or coughing are non-specific signs of many different clinical diagnoses. Nuisance barking does not necessarily indicate that the dog is experiencing either psychological or physical disturbance, but does not rule out those causes either. Differential diagnoses for barking can include: Normal motivations- play, attention seeking, excitement, frustration, social facilitation, Territorial behavior, Fear behavior. Signs of a behavior disorder- Noise phobia, Separation anxiety, Territorial or fear aggression, Compulsive disorder, Cognitive dysfunction syndrome. 1,2

2 There are a variety of commercially available tools to decrease nuisance barking. It is important to understand how they affect the animal to decrease barking. In order to fully understand this, a basic understanding of learning theory is needed. Classical Conditioning This type of learning, best associated with Ivan Pavlov, involves making involuntary emotional or physiologic associations between two unrelated stimuli. A previously neutral stimulus attains meaning or becomes a predictor (now a conditioned stimulus) for something after it was paired with an inherently meaningful (unconditioned) stimulus. 3 This type of learning is occurring all the time, whether we intend it to or not. Examples include negative association such as with white coat syndrome or a positive association with the clicker during clicker training. It occurs in conjunction with operant conditioning, and an emotional association can be made with the situation even when the appropriate operant response is made by the individual. Operant Conditioning- This is also known as trial and error or instrumental learning, where an individual makes associations between stimuli and responses. The individual learns that a particular behavior has a particular consequence. In order to understand operant conditioning, one must think about the Law of Effect. 4 This law states: Behavior resulting in pleasant consequence strengthened/ increases in frequency Behavior resulting in no/ neutral consequence is weakened/ decreases in frequency Behavior resulting in unpleasant consequence weakened/ decreases in frequency One key to this law is that the interpretation of the consequence based on pet s perception, not ours. Though we may intend the consequence of an action to be pleasant, neutral or unpleasant, if the animal does not perceive it as so, the behavior will not change as intended. There are 4 basic quadrants of operant conditioning: 5 One way to make these terms more understandable is to look at the definition of each term in the phrase then add them together again: Positive reinforcement- add something to increase the behavior. This usually ends up being something pleasant that is added. Example: food treat given after a sit. Negative punishment- remove something to decrease the behavior. This usually ends up being something pleasant/ desired that is removed. Example: turn back (remove attention) when dog jumps up. This is often combined with positive reinforcement; once the pet offers more appropriate behavior (feet on floor), the pet is rewarded. Negative reinforcement- remove something to increase the behavior. This usually ends up being something unpleasant or aversive is removed. Example: release pressure on choke chain when dog sits. Positive punishment- add something to decrease the behavior. This usually ends up being something unpleasant or aversive added. Example: choke collar pop when dog pulls ahead on leash, shock applied during barking. It is important to know difference between these in order to understand what is happening to the animal when using different tools or techniques to change behavior. Welfare Implications of Operant Quadrants with Examples First a few definitions:

3 An electronic shock collar delivers an electric shock of varying intensity either by a hand held remote or by a sensor that picks up barking. These have also been commonly called E-touch, E-stim, or static collars. They all work by shock, regardless of the name. A citronella spray collar delivers a spray of citronella from the collar directed at the dog s chin triggered by a microphone sensor on the collar that picks up barking. A scentless spray collar is similar to a citronella spray collar with the exception that there is no scent included with the puff of air. All three of these collars are intended to be used as a form of positive punishment for barking. The dog barks, the collar activates, and the dog barks less in the future. This section will focus on positive punishment because as noted above, this is the technique most often intended to be utilized when applying fast acting tools to decrease barking. To use operant conditioning effectively, but particularly positive punishment, certain criteria must be met. This is a complex technique that is difficult to use and therefore often ineffective in practice. If it is going to be effective to change behavior, repeated repetitions will not be required. If several repetitions are required, it is not working, likely due to incorrect application, and attempts to use it should be stopped. Motivation strength- The strength of the motivation to avoid the punishment must be greater than the motivation to perform the unwanted behavior. Whether this criteria can be met depends completely on the dog and how strong the motivation to bark is. A dog highly motivated to bark may continue to bark through very high levels of physical pain. Contingency- The punishment must occur every single time the behavior occurs and never when it doesn t. There are some reports, especially with the spray collars, that the microphone can be triggered by other external noises and the dog is sprayed even when he did not bark. This results in the dog not knowing why he is being sprayed, increasing anxiety, confusion and stress. This is a way that a negative classically conditioned association can be formed. Intensity- the intensity of the punishment must be strong enough to inhibit the behavior but not so strong that it traumatizes the animal. Whether this occurs again is entirely dependent on the individual dog. What is traumatizing for one dog might be a slight annoyance or just an interrupter for another dog. Often there is a fine line between aversive enough to stop a behavior and traumatic. Once that line is crossed, often the damage is already done. A classically conditioned negative response has been made. The dog cowers when the collar is put on, the dog refuses to go in the yard where he was shocked. If the dog considers it only a mild interrupter, repeated application will not decrease the behavior, and the dog will habituate to its presence. Timing- the punishment must be applied as the behavior is occurring and no later than ½ second after it starts. If timing is off, use of these techniques turns into a random act of violence against the animal. The latency to deliver a shock can be variable depending on the brand and quality of the shock collars. Particularly those dependent on an operator to push the button on the remote, if there is a delay, the dog could get shocked well after the behavior occurred, resulting the dog not knowing why he is being sprayed, increasing anxiety, confusion and stress from random application from the dog s perspective. Alternative-the animal must have a way to avoid the punishment and to stop it. This means the animal must also learn an alternative behavior.

4 There are many different side effects that can occur when using aversive conditioning or techniques even when trying to modify normal behavior. Positive punishment doesn t teach the animal an appropriate behavior, it just tells it what not to do. This can lead to increased frustration. The motivation to bark is still present, but the animal is thwarted from using that behavior as a release. Aversive techniques inhibit learning. The animal no longer offers behaviors. How can you humanely teach an animal what you want him to do if he never offers any options? These techniques can create or worsen anxiety related problems. If you consider how these techniques work to decrease a behavior, the individual is afraid of the consequence and therefore less likely to repeat it. If one is trying to modify a behavior that is originally motivated out of fear or anxiety, it does not address the motivation for problem behavior. At best, the behavior might be temporarily inhibited, but this becomes inhumane when it is at the sake of increasing an emotional state associated with poor welfare. This is an important reason why it is inappropriate to use aversive tools to try to modify fear or anxiety based behaviors. Unintended classically conditioned associations can be made. The animal may begin to fear the situation, unintended stimulus, or person associated with the punishment. When fear and anxiety are increased, many behavior problems worsen. This becomes dangerous for those around the dog because a more intense level of aggression may be displayed, but also becomes more dangerous for the dog. If he begins to panic more from separation anxiety when shocked for barking, he may become destructive in an attempt to escape, causing property damage and self-inflicted injury. The use of aversive techniques requires the handler to watch for and focus on bad behavior. This damages the human- animal bond. Do They Work? Are they Humane? Juarbe-Diaz and Houpt 6 compared electric shock and citronella collars for nuisance barking. They found that barking decreased with both collar types but it decreased more with the citronella collar (25% with shock collar and 77.8% with citronella) and owner satisfaction was higher (88.9 %) with that style than the shock collar (44.8%). They noted that some dogs cried out but still barked through the shock and some of the citronella collar sensitivities had to be adjusted to prevent spraying from ambient noises. 6 Moffat and Landsberg 7 compared citronella and scentless spray collars for barking in a veterinary kennel. Both groups decreased barking though the citronella collar performed a little better (76.7% with citronella, 58.6% with scentless). They noted a few dogs had increased anxiety with collars, and one of these dogs was diagnosed with separation anxiety after the study. 7 Deborah Wells 8 compared dogs wearing a citronella collar continuously versus intermittently and found both groups showed a decrease in barking but maintenance of lower barking rates was noted with intermittent use. All dogs showed increase in frequency of barking over time, indicating habituation to the citronella. 8 This indicates that the citronella may act as more of an interrupter than a positive punishment. All of these studies used criteria to exclude dogs barking due to behavioral pathology. 6,7,8 This was because each of these authors recognized it is inappropriate to use aversive methods (positive punishment) to try to reduce an anxiety motivated behavior. One of biggest problems arises when the general public has access to these tools to control barking when the motivation for the behavior has not been investigated. If they have not been

5 counseled on appropriate use or side effects, they often unknowingly inflict pain, fear and distress on their beloved pet. If the problem barking is due to fear and anxiety, these tools are inappropriate and quickly becomes inhumane for the pet. Potentially even worse is when these tools are recommended by public service municipalities as a way for the owner to avoid punishment (monetary fines, eviction, etc.) due to their pet s barking. It is offered as a quick fix with no attempt to help the owner understand why the dog is barking or to modify the environment or triggers leading to the barking, a more humane approach with longer term effective outcome. Though there are many veterinarians with interest in behavior, qualified trainers and also an increasing number of veterinary behavior specialists available to help owners work through this approach, these are often not offered to the owner. Some owners will seek out immediate and permanent methods to eliminate unwanted barking. The AVMA statement on debarking surgery for dogs 9 points out that this is not without its risks, failures and complications. If not enough of the vocal cord tissue is removed, barking could resume at normal volume. The surgery itself can lead to scar tissue in the airway or stenosis, resulting in respiratory compromise. Recently, there have been some newer technologic tools to arrive on the market that apply some of the other quadrants of operant conditioning to help decrease barking in a more humane manner. These may appeal to those people that want to change behavior rapidly, to those that enjoy using high tech gadgets, and to those who wish to avoid hurting or frightening their dog. The AutoTrainer 10 is a bark activated collar cued to a remote treat dispensing device. This utilizes positive reinforcement to decrease barking (actually rewarding silence). The collar identifies the bark, then starts to reward the dog with food as soon as increment of silence is measured. The dog is reward for longer and longer durations of quiet behavior. As with other bark activated collars, the handler does not need to be present. The Manners Minder 11, also known as Treat N Train is a remote controlled treat dispensing machine. This tool can be used to teach alternate behaviors using positive reinforcement. The handler can be a distance away or in the other room and still activate the machine. It also has an automatic setting. This does require some handler-dog training sessions. The Pet Tutor 12 is another remote controlled treat dispensing machine that combines many of the same options as the Manners Minder and Auto Trainer. It can be hooked to a computer for remote monitoring and control. None of these tools are appropriate alone to treat barking due to an anxiety related behavior disorder. As stated by the AVMA, Identification of the underlying cause of excessive barking is essential so that targeted therapies can be used and requires a carefully taken behavioral history. Corrective techniques may include environmental manipulation, behavioral modification, medication or combinations of these therapies. 9 Rather than first reaching for aversive tools when presented with problem barking, emphasis should be placed on creating a list of qualified individuals that are willing to counsel owners and help them work through the following steps. Guidelines on choosing an appropriate and humane plan to change behavior: What is the problem behavior?

6 What is the motivation for the problem behavior? Is this a normal behavior or a sign of a bigger behavior problem? What does a satisfactory result look like? (What do you want the animal to do instead?) What conditions can be applied to the situation to get the satisfactory result? Environmental management, behavioral counseling with a qualified professional. Positive or negative reinforcement, positive or negative punishment Then look at each option: Is it humane? Are there side effects? Choose the most humane method, apply plan. Reassess progress. Summary 1. Barking can be a normal behavior or a sign of a serious behavior disorder. Regardless of the cause, it is often considered socially unacceptable and owners are pressured to make this behavior stop. 2. There are many tools available intended to decrease barking. These utilize a variety of operant conditions to accomplish this, including positive punishment and positive reinforcement. 3. Positive punishment (and negative reinforcement) have potentially severe side effects on animal behavior and are inhumane and ineffective methods to change behavior motivated out of fear or anxiety. 4. Identifying the motivation for the barking must always be the first step in addressing problem barking, regardless of whether due to a normal motivation or behavior disorder. The plan should be tailored to that pet s motivation, environment and situation. 5. The welfare implications of any behavior plan should be considered prior to implementing the plan. References/Suggested Reading 1. Cross, NK, Rosenthal, K., Phillips, CJC. Risk factors for nuisance barking in dogs. Aust Vet J 2009; 87: Flint, EL. The function, social implications and management of barking in dogs. CAB Rev 2012; doi: /pavsnnr Houpt, KA. Learning. In: Houpt, KA, ed. Domestic Animal Behavior for Veterinarians and Animal Scientists 5 th ed. Wiley-Blackwell, Ames. 2011; Thorndike EL. The law of effect. Am J Psych 1927; 39: Skinner, BF. Operant Behavior. In: Skinner, BF, ed. Science and Human Behavior. The Free Pres,s New York. 1953: Juarbe-Diaz, SV, Houpt, KA. Comparison of two antibarking collars for treatment of nuisance barking. J Am Anim Hosp Assoc 1996; 32: Moffat, KS, Landsberg, GM, Beaudet, R. Effectiveness and comparison of citronella and scentless spray bark collars for the control of barking in a veterinary hospital setting. J Am Anim Hosp Assoc 2003; 39: Wells, DL. The effectiveness of a citronella spray collar in reducing certain forms of barking in dogs. Appl Anim Behav Sci 2001; 73:

7 9. AVMA. Welfare implications of canine barking. Jan 15, Canine%20Devocalization-Final.pdf 10. Auto Trainer Manners Minder/ Treat N Train Pet Tutor.