Schilder Comments

H

ere are some comments about one of the most frequently quoted anti-Ecollar studies. It's the one done by Schilder and van der Borg, and is called "Training Dogs With the Help of the Shock Collar: Short and Long Term Behavioral Effects". The full study is attached at the end of this article.

Comments from the study are in red. My comments are in black. The numbering of paragraphs in this article is from the original study. 

1. Introduction: A large variety of training methods are currently being used in dog training. These run from very "friendly" methods (e.g. clicker training) to the use of apparently harsh methods, like beatings and use of electronic and other equipment, that could cause wounds, pain and mental harm to the animals.

The fact that they place the Ecollar into the same category as "beatings," and "other equipment" that can cause wounds, pain and mental harm to the animals, indicates that they have a strong bias against them. Ecollars are not capable of inflicting such injuries. Having such a bias ensures that the study will show whatever they want it to show. Obviously they were not talking about modern use of the tool, at low levels of stim.

For many decades the electric collar was being used in the Netherlands, especially in training of police and guard dogs and for hunting and rescue work.

The Netherlands is well known throughout the dog training world for using the highest levels of stim possible on whatever brand or model of Ecollar available. And since it's always worked for them, they're not interested in changing. One problem, is that this part of the world is also well known for dogs that are very handler tough, often to the point of handler aggression. They regard this as "not a bad thing."

… in general a current of a few thousand volts is used.

Here's something that I find quite interesting about this study. They go into great detail about the equipment that they used. We know the various breeds of dogs used. We know their sexes and ages. We know how many wore Ecollars and how many did not. We know the brand of camera used to film the study; its model number and the size of film it used. We even know that it had a 40X optical zoom! We know the sampling method they used, the number of training sessions they observed, and the number of sequences they filmed. We know the OB commands that were used during the "walking" phase of the study, and what "protection" movements were involved. We know how the data was analyzed; we know what sampling method was used; we know how each ear and tail position was scored, and we know how the data from the two samples was compared.

But nowhere does it state what brands or models of Ecollar was used! I find this omission startling! That is, until I came across the lie that is told in the study, that in general a current of a few thousand volts is used. When this lie was told about Innotek Ecollars in Australia by the RSPCA, it cost them tens of thousands of dollars.  The RSPCA was sued by Innotek for making that statement. In this study, the statements regarding a few thousand volts would have these people being sued if they had mentioned the name of the Ecollar that was used. It's an outright lie. They avoided such a lawsuit by not mentioning the brand name of the Ecollar that was used.  While it's certainly a more powerful unit than most, it's still nowhere near "several thousand volts." 

It turns out that they used several brands of Ecollar. Most of them were probably the obsolete Shecker Teletakts, known as the most powerful Ecollar made. This model has contact points located on opposite sides of the dog's neck, (as opposed to most modern Ecollars that have their contact points about 1 1/4" apart), so there's MUCH MORE tissue involved than with modern Ecollars.

In any case, this statement is complete nonsense, and places anything said later in doubt. This "few thousand volts" figure, can only be measured on a test bench. It's called a "no load" measurement, and is useful ONLY for testing the equipment. It has nothing to do with how much voltage is in use when the Ecollar is on the dog. When training, the Ecollar contacts are against the dog's skin, resistance to the voltage is always present, and the actual voltage delivered to the dog varies from about three to about two-hundred volts. It's apparent that these authors had an agenda, since they've somehow overlooked this very simple fact - and NOWHERE in their study do they mention this fact. They leave the reader to believe that the dogs receive thousands of volts!. A scare tactic at its worst, since this purports to be a scientific study.  

Shocks are used mostly as a punishment, although some manufacturers promote the electric collar by stating that it should be used as a negative reinforcer

Few of the trainers in this study use the tool as a negative reinforcer. It's almost always used as positive punishment.

. . . no systematic investigations regarding possible long term effects of the use of the collar have been published.

I don't know where these folks have been since the late 60's when Ecollars were invented, but there have been many studies that tried to show this, most of them done by people who oppose the use of the Ecollar. Not one study has ever shown any detrimental long-term effects. If there had been even the slightest of ill effects, then those opposed to the Ecollar would be shouting them to the heavens. This study, like all that preceded it, failed to show any harmful long-term-effects from the use of the Ecollar.

Only one study (Beerda et al., 1997) showed behavioural and cortisol effects upon the reception of some shocks in laboratory dogs, that suggests that shocks are unpleasant.

Do we really need laboratory tests to know that shocks at the highest levels of stim that are available are "unpleasant?"

. . . dogs in guard and police dog training are of another breed and thirdly, they are very excited during training. This could well influence pain sensitivity.

Such a statement, that "excite(ment) . . . could well influence pain sensitivity," displays an incredible lack of knowledge of training dogs. Even the average pet owner is aware of this.

We were interested especially in finding occurrences of pain, fear, avoidance, pain-induced aggression and submission.

Ya think there's an agenda here?

2. Materials and methods: We studied the direct behavioural reactions upon administration of shocks to 15 dogs that were trained for the official (IPO) certificate as police service dogs.

People training for this certificate in this part of the world, are among those who commonly use very high levels of stim.

We also were allowed to study 31 more dogs that followed standard watchdog training for a comparable (VH3) certificate, the highest possible in this type of training.

The term "watchdog" there, has a very different connotation than in the US. IPO is one of the more difficult certifications to obtain. It requires a very high degree of control of and aggression from the animal. 

Although shocks may be painful, this does not imply that there is physical damage. A recent report on possible damage by the use of shock collars provides no evidence for physical damage and states that this is even unlikely.

No study has EVER shown any physical damage from the electrical current produced by an Ecollar. For the most part, those opposed to the use of the Ecollar have given up any hope of showing this.

The second point illustrates that effects of the electric collar, at least when used in a harsh way, may be visible outside the training area.

It would seem that there is some acknowledgement of the existence of low-level stim use here. But it's not part of the study.

We hope a future comparison of German shepherd dogs trained in a more friendly way will bear out that indeed a friendly training regime leads to less signals of stress. We have not proved that the long-term welfare of the shocked dogs is hampered, but we have made clear that it is under serious threat.

The researchers admit, "We have not proved that the long term welfare of the shocked dogs is hampered." I wonder if they'll include low-level stim in those tests of "more friendly way(s)?"

4.4. Why is there so much and such heavy punishment during police and guard dog training? First, this type of training is typically and traditionally work by and for men: it is mostly men, that do these trainings and they have been doing it their way and successfully for many years. Men mostly are harder on animals than women, men may be perceived as more threatening than females

They cite no evidence for this sexist opinion.

Secondly, training time is too short.

Too short? Time is relative of course; but with some so-called "kinder, gentler method" trainers taking years to get one simple behavior, yes, training time is too short. IPO is a complex test requiring much from the dog and the handler.

Thirdly, prestige is an important factor: championships or high rankings count heavily. All this promotes severe punishment in order to get quick results.

What one person may think is "severe," another will think is commonplace. The term "severe" is never defined.

Success in training does not promote willingness to change the type of training.

"If it ain't broke, don't fix it."

Fourth, the type of dog used here are highly motivated, hard, temperamental and possess low biting thresholds. They stem from special breeding lines and quickly become so excited, that mistakes are easily made and commands not obeyed. This also promotes punishments.

These dogs have been bred like this for decades. A change in the training method is not likely to occur anytime soon. Using dogs bred like this may take a bit more time to produce a top dog using low stim levels, but we'll have to wait and see.

Also, the high excitement may have analgesic effects, so that softer physical punishments do not have the desired effect.

"The high excitement MAY have analgesic effects?" These people may know their stuff when measuring and computing data, but it's fairly obvious that they're not dog trainers.

Fifth, during training excessive emphasis is being laid upon biting. The "let go" command, that is often not obeyed and therefore elicits punishment, is trained much later than biting.

It's silly to say that the emphasis laid upon biting is "excessive." This is a biting sport, and of course the emphasis is on biting. I think it's a mistake to train a bite and not to train an out, particularly in the litigious society that the US has become. But the Netherlands does not have this issue. Many trainers think that introducing the out command weakens the bite, and in their training system, the way that the out is trained, it probably does.

Last, these dogs are expensive to extremely expensive animals. Handlers and trainers both want to get the maximum out of the animals.

Dog owners worldwide, no matter how much their dogs cost, want to get the maximum out of them. If these trainers thought that the Ecollar was doing damage, they certainly would not be using them.  

4.5. Conclusions and recommendations: We concluded that shocks received during training are not only unpleasant but also painful and frightening. Furthermore, we found that shocked dogs are more stressful on the training grounds than controls, but also in a park. This implies, that whenever the handler is around, the dog seems to expect an aversive event to occur. A second unwanted association might be that the dogs have learned to associate a specific command with getting a shock.

Keep in mind that this study ONLY dealt with the highest levels of stim possible from one of the "hottest" Ecollars available.

It's interesting to note that during the testing, the "control group" wasn't wearing Ecollars, and so the researchers could easily tell which dogs were getting stims. It's obvious that they could slant the study any way that they wanted. To have any value, both the control group, and the dogs receiving stims should have been wearing Ecollars. Because of this lack of a "blind control group" (one where the researchers didn't know which dogs were getting stimmed and which ones weren't) their comments about the dogs are meaningless.

Apart from the acute pain and fear, these expectations may influence the dog's well being in the long term in a negative way. To counter misuse of the shock collar, it is proposed to ban its use for "sports", but save it for therapeutic applications, such as for suppressing hunting and killing sheep.

Here we can see the agenda.

This study used highly subjective measures: it is obvious that the researchers did a poor job of investigating without bias, used flawed methods to measure "fear," and drew worthless conclusions for the way Ecollars are used with modern methods today. 

One of the things they measured was the "squeels" (sic) emitted by the dog when he received the stim. If a handler had stepped on a dog's foot, it probably would have squealed, but not from being stimmed. Yet the "researchers" would have recorded it as if an Estim was responsible. Stepping on a dog's foot, is much more likely to draw a fear response from a dog than an Estim.

They measured ear carriage as an indicator of fear. How do they know that the dogs were afraid of the stim? How do they know that it wasn't another handler, another dog, a decoy, something else in the environment that was causing it? Truth is, they didn't. They guessed! And since there are many reasons that a dog will have an ear carriage that indicates fear, this way to measure it is highly flawed.

Following are some comments from Steven Lindsay.  They are in blue. Lindsay is a dog behaviorist and trainer who provides a variety of behavioral and counseling services. He previously trained and evaluated military working dogs as a member of the US Army Biosensor Research Program, commonly known as the Superdog Program. He also does seminars and workshops. He wrote the three-volume set of books called "Handbook of Applied Dog Behavior and Training."

Lindsay writes (regarding this study specifically) The main differences observed between dogs receiving ES (Electrical Stim) and those receiving other forms of correction included an altered ear posture detected during obedience work and free walking, tongue flicking (appeasement licking) during protection work, and submissive pawing actions during obedience work.

These observations can be the result of many other things besides "fear," which is what the "observers" noted. Yet they only reported that one result, and never remarked that any other stimuli could be responsible. The "altered ear posture" is completely subjective, and these anti-Ecollar people could have easily interpreted the ear carriage any way that they wanted to. The same criticism applies to tongue-flicking and pawing actions, other behaviors that they measured.

Lindsay writes, The notion that Dutch working dogs might have become fearful of their handlers as a result of shocks received in training is reported as an obvious fact that is never actually tested, leaving it to the reader to accept the speculation "as fact" or not. In practice, dogs do not appear to link ES with the handler, especially persons with whom the dog is closely attached and familiar. In fact, the most interesting uses of the collar depend on this lack of aversive association, including lasting reward and opponent safety effects (Denny, 1991). Interestingly, the IPO system has devised a good behavior test for detecting mishandling and abuse. Surely, if an IPO dog had developed a fear or aversion towards its handler as the result of electrical training, the following IPO Watchdog Test [WH (Wachhunde Certificate)] requirement would likely reveal it, causing a great many dogs to fail if they were treated as badly as alleged by the present report. 

If the electrical and physical stimulation were truly traumatic and stressful, one would expect that the traumatized dog might be apt to flee at the first instant it got a chance. Further, one would expect that its willingness to bite and hold the sleeve ought to decrease in proportion to the amount of fear and pain it experienced (e.g., causing the dog to come off the sleeve too early or not to bite as hard) or that the dog might even show signs of avoidance and fear toward the agitator. However, no such loss of drive or performance is reported. In fact, Dutch dogs are renowned for their hardness, work enthusiasm, and acrobatic attacks - attributes that are opposite to what one would expect from training that was overly stressful.

With increased biological stress, as in sickness, one would expect to observe a drive-reducing effect on aggression and a loss of voluntary initiative, whereas increased fear should tend to suppress behavior rather than enhance it. The absence of reduced drive or behavioral suppression with respect to critical activities associated with shock (e.g. bite work) makes one skeptical about the lasting adverse effects the authors claim to document. Although they offer no substantive evidence of trauma or harm to dogs, they provide loads of speculation, anecdotes, insinuations of gender and educational inadequacies, and derogatory comments regarding the motivation and competence of IPO trainers in its place.

Most scientific evidence supports the notion that the cessation of aversive ES in the context of escape/avoidance training is more likely to enhance social attraction, promote feelings of safety, and calm a dog rather than make a dog afraid or apprehensive. These secondary effects of shock termination and pain reduction have long been recognized to promote conditioned and unconditioned effects conducive to reward and safety (see "Electrical Stimulation Controllability and Safety"). Instead of instilling social aversion and anxiety as suggested by the authors, competent electronic training may actually promote social attachment, reward, and safety. With the behavior- contingent cessation or avoidance of ES, dogs experience immediate emotional relief that subsequently merges into a state of progressive relaxation incompatible with social aversion and fear - a sequence of opponent emotional effects contrary to those alleged to occur in the case of working dogs exposed to ES in the context of training.

. . . once critical limits and fair-play rules are set via inhibitory conditioning (e.g. all-stop, stop-change, and go/no go), the social and play rewards associated with the activity itself should exert potent counter conditioning effects, thereby further reducing any secondary aversive emotional conditioning effects arising from inhibitory conditioning with shock. These various bits of circumstantial evidence conflict with the allegations that shock, … promotes social fear, stress, and represents a serious threat to the long-term welfare of trained dogs. (Bold Emphasis added)

Trainers and handlers should study learning theory far better and review the structure of the training in order to teach the let go command in an earlier phase and to reduce the number of mistakes. They should incorporate more rewards during exercises. Also, less temperamental and less forceful dogs should be bred. This also would decrease the chance that dogs make mistakes for which they could receive punishment.

It's interesting how a supposedly "scientific" study turns into something like this. I have no doubt that they began with their conclusion and worked towards it. That means that all data came through that filter. A couple of quotations come to mind.

"It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories instead of theories to suit facts." -- Sherlock Holmes (Sir Arthur Conan Doyle).

"There is a principle which is a bar against all information, which is proof against all arguments, and which cannot fail to keep a man in everlasting ignorance. That principle is contempt prior to investigation."
-- Herbert Spencer.