Reward-Based Dog Training: Why the Evidence Matters

If you’ve ever wondered why we only use reward-based dog training methods at Second Nature Behaviour, the short answer is simple: the science tells us it works — and it’s better for your dog. But there’s a fascinating story behind that answer, and understanding it can change how you think about training altogether.

Whether you share your life with a dog, cat, or parrot, the principles of how animals learn are remarkably similar. So let’s take a look at what reward-based, evidence-based dog training actually means — and why it matters.

What Is Reward-Based Dog Training? 

Reward-based dog training means using things your dog or pet finds motivating — usually food, but also toys, games, praise, or experiences — to encourage the behaviours you want. We’re essentially making a behaviour more likely to happen again by rewarding it, while simply withholding the reward for behaviours we don’t want. Think of it as paying your dog for a job well done, much like you getting paid at work.

The beauty of this approach is that it taps into your dog’s natural motivation. Different animals find different things rewarding, and even within a species, preferences vary hugely. Some dogs respond well to verbal praise, some love physical affection, while others couldn’t care less about a head scratch but would do anything for a piece of cheese.

Finding What Motivates Your Dog

It’s worth taking the time to discover what your pet truly values. For dogs, food rewards usually top the list, but there’s a hierarchy even within that — kibble is typically low on the list, with pâté and chicken higher up, and cheese often taking the number one spot. For cats, a feather toy might be more motivating than a treat. For parrots, granivores may go wild for sunflower seeds or pine nuts, while nectarivores may prefer fruit or honey.

Try offering a variety of rewards and changing them regularly to maintain your pet’s interest. We mainly use food in training because it can be delivered quickly and efficiently, and research has found food to be the most effective reward in the early stages of learning. But food isn’t needed forever — it’s gradually phased out and replaced by other rewards as training progresses. Your behaviourist will guide you on this.

The Science Behind How Animals Learn

The methods we use are grounded in decades of scientific research. Two key principles underpin everything we do.

Operant Conditioning

Developed by B.F. Skinner and built on Thorndike’s Law of Effect, operant conditioning is based on a straightforward idea: behaviours followed by pleasant consequences are more likely to be repeated, and behaviours followed by unpleasant consequences are less likely to be repeated.

In practice, this means we teach an animal to perform a behaviour and reward them quickly with food. The next time we ask for that behaviour, the animal is more likely to offer it because they’re anticipating the reward. If they don’t perform the behaviour, we simply withhold the treat and ask again — or go back a step to help them understand. This is how reward-based dog training works in everyday terms: rewarding what we want and calmly redirecting when things don’t go to plan.

Classical Conditioning

You may have heard of Pavlov’s dogs from school — that’s classical conditioning, which is learning through association. We use this approach to change how an animal feels about something or a situation.

For example, a dog who is afraid of having their nails clipped can learn to associate the nail clippers with something positive. Each time the clippers appear, a tasty treat follows. Over time, the dog begins to feel differently about the clippers — they become a predictor of good things rather than something to fear. This is a simplified account, of course; in reality, a careful plan is needed, working at the dog’s pace.

What the Law Says

In England and Wales, the Animal Welfare Act 2006 requires anyone responsible for an animal to meet its welfare needs, including protecting it from pain, suffering, injury and disease. The Code of Practice for the Welfare of Dogs (Defra) describes reward-based training as the preferred approach, and warns that training involving physical punishment can cause pain, suffering and distress and can worsen the very problems it aims to address.

Professional codes of conduct apply too. The ABTC Code of Professional Conduct requires practitioners registered with the Animal Behaviour and Training Council (ABTC) to use science-led, non-punitive methods, and bars them from endorsing “correction” devices such as prong collars, check chains and electronic training collars. CCAB Certification Ltd’s Code of Conduct allows a Certificated Clinical Animal Behaviourist to be suspended or removed from the register for misconduct.

What the Research Shows

The evidence supporting reward-based dog training is substantial and growing. Here are some of the key findings:

Research has consistently found that dogs trained using reward-based methods perform better when learning new tasks and are more attentive to their owners. Studies have also shown that reward-based methods are associated with fewer reported behaviour problems, and that punishment-based approaches can worsen anxiety, fear, and frustration — leading to more problem behaviours, not fewer. Owners who used only reward-based methods reported more responsive, cooperative dogs.

Perhaps most strikingly, dogs trained using aversive methods have been found to have a more pessimistic outlook, suggesting a more negative underlying mood state. Research has also shown that owners who use confrontational methods — such as lead corrections, verbal punishment, or physical interventions — often reported that their dogs responded aggressively.

Studies on electric shock collars specifically found that their use increased distress in dogs and did not improve training outcomes compared to reward-based alternatives. Electric shock collars have been banned in Wales since 2010, and are banned or effectively banned in several European countries, including Austria, Switzerland and Germany. Sadly, they are not yet banned in England, Scotland or Northern Ireland.

Why It Matters for Your Pet

When you choose reward-based dog training, you’re choosing an approach that builds trust between you and your pet. Training becomes something enjoyable — a shared activity rather than a battle of wills. Your pet learns faster and develops genuine confidence rather than suppressed fear.

Whether it’s a dog learning to walk calmly on a lead, a cat becoming more comfortable with handling, or a parrot learning to step up, reward-based methods lead to a happier, better-behaved pet and a stronger bond between you.

If you’d like support with your pet’s behaviour or training using a reward-based approach, get in touch today to see how we can help.

If you need help with your pet’s behaviour, get in touch with Second Nature Behaviour today. We’d love to help.