The Training Program

THE TRAINING PROGRAM

OUR PROGRAM

Our training program was developed by expert dog trainer Seth Pywell. Seth is widely considered Western Australia’s best dog trainer and has been training and reconditioning dogs for almost 20 years, specialising in canine behaviour modification, scent detection, protection training and obedience training. The program was originally developed in the late 90’s for Working & Hunting dogs, then later modified to suit Search & Rescue Dogs. Such dogs by their very nature are extremely forward going and inquisitive of new environments (Good nerves and well driven). The program has been continually refined over the years to ensure it uses up-to-date, scientifically sound methods which maximise your dogs’ potential for learning. Field testing of the training has also been used to eliminate any weaknesses and produce the most reliable snake avoidance training system available today.

Live snakes MUST be used during Snake Avoidance training. The majority of training principles for snake avoidance are the same as those that govern reliable scent detection training. Snake faeces, shed skins and/or toy snakes will not teach your dog to avoid real snakes, they will only teach the dog to avoid those specific items which is of no use. You must remember that a human can see a toy snake and get scared because we first understand what a snake is and that engaging it can bring an unpleasant consequence. Dogs do not know what snakes are which is why they must investigate them in order to learn. Bear in mind that smell is a key sense used by dogs to understand the world around them. Live snakes have a distinctly different odour to snake faeces, shed skins and toy snakes, nor do they behave the same. If humans weren’t first taught what a snake actually is, a toy snake would hold no meaning; it would be nothing more than an oddly shaped piece of plastic. By the same token, snake faeces and shed skins hold meaning to humans because we know where they originate; this is obviously not the case with a dog. It is also important to note that snakes do not create homes, they defecate and shed while moving around over large areas. Training a dog to avoid articles such as droppings and shed skins, which do not put the dog in danger, installs fears that will cause unnecessary stress. Should these articles appear on your property, they will stay there until they are physically removed or decay. While your dog may initially avoid these articles (if correctly trained to do so) over time your dog will be spurred on by its natural curiosity to investigate (instinctive drift) and eventually overcome its fears, meaning the training was a waste of time and money. A live snake will try to get away from your dog and use the time provided by the Avoidance Training to leave your property. Snakes do not want to get into confrontations; it compromises their survival by potentially leading to injury or death. It should also be mentioned that deceased snakes cannot be used because they too have a different odour to a live snake. The moment an animal dies, it begins to decay. This is why cadaver search dogs used by police must be trained using samples from actual cadavers.

Laws of generalisation state that for a dog to properly understand what it is being taught, in this case Snake Avoidance, it must be exposed to a minimum of 8 different samples in 8 different locations and 8 different situations. Reliable Snake Avoidance simply cannot be taught with less.

Due to the way dogs learn, in addition to the minimum requirements mentioned above, odour and movement avoidance must be taught separately, as well as combined. Without this a dog may be bitten while sniffing a snake hidden in long grass (no visual) or approach a snake that is downwind (no odour). To expect a dog to endure the training required to teach proper Snake Avoidance in one session would overwhelm the dog and would not be as effective as breaking up the training.

The Self Discovery aspect of this training is very important and cannot be overstated. Due to the fact that it’s impractical to monitor a dog 24 hours per day, our training has been designed to allow our dogs to learn from their own decisions. For obvious reasons Snake Avoidance training should not be dependent on the presence and interference of a human. 

Employing any Snake Avoidance training techniques that require constant interference from humans is extremely lacking and impractical to say the least.

Animal welfare has always been a core value in the development of our program, which is why both a dog trainer and snake handler are present at every session. In addition to these professionals providing firsthand safety of both the dogs and the reptiles, we also maintain a number of mandatory safety protocols followed throughout the training process.

This training is provided as a preventative measure, designed to safely educate our dogs to the consequences of engaging venomous snakes. It can be said that learning takes place when the expectation differs from the outcome. Most dogs clearly expect interacting with a snake to be somewhat interesting and enjoyable as they are generally very curious and quickly engage the snake. After learning that engaging a snake brings an unpleasant consequence, our dogs demonstrate that learning has taken place by their future decisions to avoid the snake. While no one can guarantee all the decisions your dog will ever make in the future, we draw a similarity between learning the road rules and road safety. Learning the road rules is not a guarantee that accidents will never happen, but they certainly have a dramatic effect on reducing the risks of road accidents. We recommend all dogs completing our Snake Avoidance Program undergo an annual refresher to test your dogs’ retention of the training and improve the probability of avoiding snakes in the future.

Why Traditional Methods Fail

Course creator, Seth Pywell, began exploring ‘American methods’ of snake avoidance training in the late 90’s after receiving numerous requests to ‘snake proof’ hunting dogs. At this time snake avoidance was not available in Australia. These traditional methods involve placing a snake in a cage or Perspex box and allowing the dog to approach. On investigating the ‘snake box’ the dog receives a correction via an electronic collar. While on the surface this training appears to work, once out in the field, dogs once again engage reptiles. The reason this training fails is because the dogs become fearful of a ‘picture’. This picture is of a particular snake, in a box, in a specific location. This hypothesis was confirmed by replacing the ‘snake box’ with an empty box after the initial training was provided. The dogs, despite encouragement, would not approach the empty box.

Due to the inability of this training method to transfer to the field, it was quickly abandoned. It was evident that a new system would need to be developed.

The Canine Laws of Learning

In the development of a new Snake Avoidance training program, several canine learning principals had to be taken into account. Five key laws of learning were identified as having a strong impact on training reliable snake avoidance, with the final goal to achieve generalisation of the learning (the dog avoids all snakes in all locations). These laws are: Individual Discrimination, Species Discrimination, Location Sensitivity, Contextual Learning, and Inadvertent Associations. Canine learning dictates that when a dog is taught to avoid a particular snake, the dog believes that it is required to avoid only that individual snake (Individual Discrimination) and only in the exact location that snake was encountered (Location Sensitivity). Further training with other members of that snake species (males, females, adults, juveniles) will help the dog understand that it is not just the original snake to be avoided but all members of that species (Species Discrimination). Once the dog understands all members of that species are to be avoided, the location must be moved and the training repeated. Only after this training is firmly understood in all locations can the introduction of the next snake species commence. Our program is specially designed to streamlines this process without compromising the dogs learning, maintaining reliability.

In addition to applying a correct process, the dog trainer must take several other factors into account. If the electronic training collar is fastened incorrectly, it is possible for the dog to become collar wise (Contextual Learning), i.e. the dog only avoids snakes while wearing the training collar. The occurrence of this collar wise behaviour can also be impacted by a dog’s previous experience with an electronic collar outside of the program. It is also imperative for Snake Avoidance trainers to be experience and fluent in scent detection training to account for other forms of Contextual Learning that can occur. Dogs will not always approach a snake from the same direction, meaning the snake’s odour may be blown towards the dog or away from in depending on the direction of the wind. If the dog is always trained from downwind, it will lean only to avoid snakes when it can both see and smell them. Conversely a dog may not be able to see a snake if it is hidden by vegetation, therefore training that is reliant on the dog seeing the snake is flawed. The dog trainer must also be able to accurately read the dog’s body language. Incorrect application of the electronic stimulus to result in the dog to believing the sensation it felt was caused by an alternative source (Inadvertent Associations), e.g. environmental factors, owner, handler, leash pressure, Perspex box, mesh cage, etc.

Reward Only Training

We are often asked why reliable snake avoidance training can’t be achieved with the diligent use of food rewards. Many trainers around the world manage to teach a number of different behaviours using food alone. Reward based training methods were not overlooked in the development of the program; Seth Pywell is an expert in this field. Food rewards are in fact an important part of the program and utilised far more than remote collar corrections. The reason that “reward only” training is ineffective is that rewards simply show the dog what behaviour we approve of, it does nothing to teach the dog that snakes are to be avoided and engaging them has a consequence. Withholding a food reward is ineffective because once the dog notices the movement of a live snake, it becomes a high value distraction and the dogs desire shifts away from the food being offered, focusing only on the snake. The dog has chosen to completely disregard the food being offered and withholding what the dog no longer wants is ineffective. By applying a consequence when the dog engages a snake, the dog’s desire for the reptile immediately ceases. Since the snake is no longer attractive to the dog, the dog can only now shift its focus back to the food reward.

We have also been asked about using snakes as a cue (command) for the dog to perform a behaviour that removes it from danger. The technical term for this is differential reinforcement of an incompatible behaviour. While it is possible to achieve, unfortunately it has several flaws which cannot be overcome through “reward only” methods. The dog would first need to be taught two very reliable recalls; one to the owner for when the dog is out on walks off-leash and one to a location at the home for when the dog is left in the backyard. The dog must also stay in that position until a release command is given otherwise it will just return to the snake. There is also the concern that in a real world scenario, the snake could possibly occupy an area nearby to where the dog was taught to recall to. This school of though was considered early on in the development of the training but abandoned due to its impracticality. The reason “reward only” methods are insufficient is that the dog performs the recall (snake avoidance) behaviours in order to obtain the food reward it has come to expect. The training is therefore reliant on a human always being present and providing a reward. If the dog performs the behaviour and food is not presented, the dog will abandon the behaviour in favour of investigating the snake, since the snake has never been associated with a consequence. For humans, this is like forgoing an enjoyable activity in favour of going to work, then finding out you aren’t getting paid for working; you will quickly stop work and return to the enjoyable activity. It is important to remember that the majority of dog-snake encounters occur while the dog is not being supervised.

For more in-depth reading, please download the following document.

  The hidden cruelty of food based training