Understanding Animal Training – Positive Reinforcement Techniques

April 22, 2024

Introduction to Positive Reinforcement

There are millions of animals in the UK. According to the PDSA, 53% of adults in the UK own pets, with dog and cat ownership being the highest. There are also numerous animals in zoos, safari parks, aquariums, farms and sanctuaries.

Where animals are in homes or captivity, they will require training, which means manipulating their behaviour (ScienceDirect). Training is necessary for the following reasons:

It enables owners, keepers and carers to build positive relationships with animals and provide high-quality care.
It provides animals with mental stimulation, enrichment, physical activity and the necessary skills to live successfully and happily in homes or captivity.
It enables owners, keepers and carers to look out for any issues regarding their health, safety and well-being.
It helps people handle animals safely for husbandry and veterinary purposes and reduces stress during these activities.
It helps to reduce problem behaviours so people and animals can live in harmony.
It can help animals develop self-control and self-confidence.
It can benefit research, especially when animals are in zoos and aquariums.

There are various training methods, and the choice will depend on the animal, species and breed. One method preferred by many trainers is positive reinforcement training (PRT), which means “increasing the likelihood that the behaviour will recur by adding something the animal desires” (Danks, L. 2016).

Positive reinforcement teaches and strengthens desired behaviours by training animals to create an association with rewards, such as treats, praise and toys, while ignoring or redirecting undesired behaviours. By rewarding positive behaviour in animals, they are more likely to continue with these behaviours in the future.

Positive reinforcement is a powerful and humane approach to training animals across species, whether for obedience, agility, therapy, or other purposes. It is a widely used technique by trainers, as it is effective and doesn’t involve force, coercion or punishment. This blog post explains the principles behind positive reinforcement, its benefits, and practical tips for effectively implementing these techniques.

The Science Behind Positive Reinforcement

To understand how animals learn through positive reinforcement and to get the most out of training, it is helpful to look at the science behind this technique.

Firstly, knowing how an animal’s brain works is important, as it will influence their response to training. There are reward systems in the brain, which are a group of structures responsible for motivation, desires and learning. When an animal expects or receives a reward, their brain reward system is activated and responds by releasing more of a chemical called dopamine, a neurotransmitter that makes animals, and also us humans, feel happy and positive.

As positive reinforcement is reward-based training, it triggers the release of dopamine, which motivates animals to work for things more by learning how to behave, act and think. Each time they receive a reward, it strengthens their brain connections and reinforces behaviours associated with rewards.

Not only is biology significant in reward-based training, but psychology also applies. Various psychological principles underpin positive reinforcement, such as operant conditioning and classical conditioning.

Operant (instrumental) conditioning

Positive reinforcement came from a theory of operant conditioning developed by a behavioural psychologist, B.F. Skinner in the 1930s. He created a device known as an operant conditioning chamber, lever box or Skinner box, which he used to study behaviour in rats and pigeons.

Operant conditioning is a behaviour modification technique that modifies an animal’s voluntary behaviour by the immediate consequences. Animals learn through repeated reinforcements and punishments. If their behaviour is reinforced (rewarded), they will likely repeat it. On the other hand, punishing their behaviour will decrease it.

Reinforcements can be positive or negative, for example:

Positive reinforcement – adding something positive to strengthen the desired behaviours in animals.

For example – a trainer wants to encourage a dog not to jump up on people and gives them a reward when they do not jump.

Negative reinforcement – removing something bad to strengthen the desired behaviours in animals.

For example – a trainer stops pushing a dog forcefully back and then down while they are jumping. As the dog does not like this punishment, it encourages them not to jump when the trainer stops.

Positive punishment – adding something unpleasant as a punishment to decrease undesired behaviours in animals.

For example – a dog is jumping up, and the trainer pushes them forcefully back and then down to discourage the behaviour.

Negative punishment – removing something positive as a punishment to decrease undesired behaviours in animals.

For example – as a dog jumps up for attention, the trainer turns away from the dog and ignores them to discourage them from jumping.

Positive reinforcement is always better than punishment, and we will look at some of the reasons in the next section.

Classical (Pavlovian) conditioning

Trainers should also be aware of classical conditioning, another learning technique first described by Russian physiologist Ivan Pavlov in the late 1890s. His theory of classical conditioning came from his research into dog digestion when he discovered that dogs learned to associate food with a bell sound and would salivate as a response.

Classical conditioning is where animals learn involuntary responses through associations. It involves pairing a neutral stimulus with an unconditioned stimulus to bring about the response, for example:

Neutral stimuli – these do not usually cause an animal to respond in a particular way, e.g. a sound, such as a bell, whistle or clicker.
Unconditioned stimuli – these cause involuntary reflexes and natural (unconditioned) responses where no learning is involved, e.g. salivating at food.

With repetition, animals can link the above two stimuli and modify their behaviour, which is known as a conditioned response to conditioned stimuli. Here is an example to make it a bit clearer:

Before conditioning

No response to neutral stimuli, e.g. a dog does not salivate to the sound of a bell.
Unconditioned response to unconditioned stimuli, e.g. a dog salivates at the sight of food.

During conditioning

Unconditioned response to conditioned stimulus, e.g. a bell sounds on presenting food, and the dog salivates.

After conditioning

Conditioned response to conditioned stimuli, e.g. a dog salivates when a bell sounds even when no food is present.

While positive reinforcement training is a type of operant conditioning, classical training can also be used as part of the process, as it is basic training and one of the simplest types of learning. Here is an example of this in clicker training.

A trainer starts to use clicker training as they want an animal to associate the clicker sound with a reward to achieve a desired behaviour.
At first, the animal will not have a conditioned response. However, with repetition, the animal will begin to associate the clicker sound with the reward and over time, the clicker will trigger a conditioned response – classical conditioning.
The animal starts to modify their behaviour and learn new ones voluntarily and purposefully in response to the clicker sound and to earn a reward – operant conditioning.

Benefits of Positive Reinforcement

As training has a significant impact on animals’ mental, emotional and physical welfare, it is essential to adopt techniques that will achieve the best outcomes for the animals and trainers. Positive reinforcement is a preferred technique, as it has numerous benefits, such as:

It is a humane and ethical way to train animals.
It gives animals a choice to participate in and have control over their environment, which promotes welfare and reduces fear, anxiety and stress.
It is a fun way to train, as rewards-based training can develop positive bonds between animals and trainers.
It builds trust, respect and confidence between trainers and animals.
It encourages animals to think for themselves.
It promotes good behaviour overall and reduces the risk of behavioural problems in the future.

Some still think that punishment is an effective way to train animals. Punishments are actions taken to reduce the likelihood of a behaviour recurring. Some examples are:

Jerking collars and leads.
Verbal intimidation, e.g. shouting, growling or angry voice.
Hitting.
Nudging.
Kicking.
Poking
Electric shock collars – a ban is still being considered in the UK.
Ignoring.

Positive reinforcement is a much better option, rather than using force, coercion or punishment to train animals. If trainers use punishment, it can result in many negative effects, such as:

Physical injuries.
An increase in undesired behaviours. A study by Hiby et al. (2004) found that punishment was associated with an increased incidence of problematic behaviours.
Animals becoming confused and not understanding the behaviours expected.
Stress, fear and anxiousness, leading to long-term effects.
Lower motivation and an increase in avoidance behaviours.
An increase in aggressive responses, potentially resulting in harm to people and other animals.
Animals associating punishment with other events.

A review by Ziv, G. (2017) into the effects of using aversive training methods in dogs concluded that those working with or handling dogs should rely on positive reinforcement methods and avoid using positive punishment and negative reinforcement as much as possible.

There are so many other training techniques that do not require punishment and are better for the health, safety, well-being and welfare of animals. Positive reinforcement is one of the best methods as it achieves long-lasting results and is a more effective, gentle, safer and caring way of training animals.

Understanding Rewards and Reinforcers

Positive reinforcement is reward-based training, but what exactly are rewards? They are incentives to motivate animals to engage in desired behaviours. Rewards should be something animals enjoy and want to work for. Some examples include:

Food and treats.
Play, i.e. favourite toys and games.
Exercise.
Attention and petting.
Praise.

Animal training relies on effective reinforcers. These are rewards that are likely to reinforce positive behaviour in animals, and there are two types:

Primary reinforcers – these are innate and natural and do not have to be learned by animals, as their biological need makes them desire it, e.g. food, sleep, play and attention. These are not dependent on other reinforcers.
Secondary reinforcers – these are stimuli that reinforce positive behaviours in animals once they learn to associate them with primary reinforcers. Stimuli may include petting, praise, sounds (i.e. clickers and whistles), hand gestures, etc. For example, a trainer gives a dog a treat (primary reinforcer) whilst using a clicker (secondary reinforcer).

For training to be effective, it is essential to select the most appropriate reward (reinforcer) which an animal desires or is interested in. Rewards should also be appropriate for the animal species, breed, individual preferences and the environment, for example:

Animal species – a dog’s reinforcers will differ from those of a cat, bird, rat or exotic animal.
Breed – in domestic animals, each breed will have different traits. For example, some dog breeds are strongly motivated by food, e.g. Labradors, and others may desire play and toys, such as Border Collies who like to chase balls.
Individual – animals have their own characteristics, personalities, underlying physical/mental health conditions and needs, and will have favourite rewards. One dog may love sausage treats, and another may prefer biscuits.
Environment – an animal’s environment will influence their preference for rewards, e.g., a domestic pet in a home will have a different environment to an animal in a zoo, which will influence reward choice.

To determine the most effective rewards, trainers must carefully observe the animal’s behaviour after giving a specific reward. If the animal is showing no interest or is not showing the desired behaviour, it could be that they don’t want to work for that particular reward. Trainers should try various rewards until they find those that are preferred.

It is important to note that animal’s preferences for rewards may change. They may get fed up with the same treat, toy or game. Any changes to their physical and mental health and environment can also influence their preferences. Keeping a record of rewards that animals respond to can help trainers select the best ones in the future. Here is some guidance on choosing different dog rewards from the Dogs Trust.

Training Techniques and Strategies

Positive reinforcement aims to increase the chances of animals behaving in a specific way desired by trainers. Here are some practical tips to maximise the chances of success:

Have the right rewards – for the reasons we have already covered. Positive reinforcement relies on rewards animals desire and want to work for.
Be realistic – a trainer’s goals should be clear, realistic, and achievable, and they should not try to do too much too soon or overcomplicate matters. Starting simply and gradually can help animals build confidence and proficiency.
Be consistent – if a trainer is inconsistent with their training approach or rewards, it can confuse animals during training, as they will not understand what is expected. Be consistent by rewarding the desired behaviour and redirecting or ignoring undesired behaviours.
Keep training sessions short – trainers should do short sessions more frequently than longer ones where animals can get exhausted. For dogs, practice for about two to five minutes at a time and five to six times a day.
Communicate clearly – for an animal to understand cues (e.g. hand gestures) and commands, trainers must communicate clearly.
Keep timing in mind – trainers must reward the animal immediately after the desired behaviour for them to make associations and to reinforce. Delays can confuse animals, as they will not know which behaviour the rewards are for.
Be patient – it takes time for animals to learn, and there may be challenges along the way. Being patient with animals, even in moments of frustration, is key to reinforcing positive behaviours.
Avoid using punishment – for the reasons we have already covered. It is not about punishing bad behaviour but reinforcing good behaviour.
Stay positive – some animals, especially dogs, can pick up on human emotions. Researchers believe that dogs can mirror the anxiety and negativity of owners. Therefore, negativity during training could have negative impacts on animals. Being positive and enthusiastic and making training fun is essential.

To help trainers get started, here are some example step-by-step guides for basic commands and behaviours, such as sit, stay, and recall:

Dogs

RSPCA – has step-by-step guides on training dogs to sit, lie down and stay.
Petplan – advises on preventing dogs from jumping and teaching recall and sitting.
Blue Cross – has some videos demonstrating how to get dogs to respond to certain commands.

Cats

Petplan – details how to train cats to come when called, sit on a person’s lap and go into a car carrier.
Cats Protection – advises on how to teach a cat to sit down, lie down and rollover.

Other animals

Blue Cross and Horse & Hound – have advice on introducing clicker training to horses.
Vital Pet Club – has information on the basics of training rats and rabbits.

Common Mistakes to Avoid

Trainers can and do make mistakes when using positive reinforcement, especially when starting, as they are learning just like the animals they are training. Here are five common mistakes and how to correct them if they occur.

Too many rewards

Positive reinforcement is reward-based training, but a common mistake people make is to overdo it with rewards or only give certain ones, such as treats.

If an animal receives too many rewards, they can become meaningless, and some may refuse to engage in the behaviour unless they receive a reward. Also, if the rewards are food-based, it can lead to obesity and health issues. On the flip side, too few rewards can make it difficult for them to understand which behaviour is being rewarded.