Why reward-based training works
Animals, like people, repeat actions that lead to good outcomes. If sitting calmly brings a treat, your dog will sit more often. Over time, the behaviour becomes a habit, and rewards can be given less often.
Research on dog training has found that reward-based methods are at least as effective as punishment-based methods and are linked to fewer behaviour problems, less stress and better relationships between dogs and owners. Punishment can suppress behaviour in the moment but often increases fear and aggression.
The key principles
- Reward what you want, rather than waiting for mistakes.
- Timing: the reward must come within a second or two of the behaviour, so your pet connects the two.
- Consistency: everyone in the household uses the same cues and rules.
- Small steps: break complex behaviours into easy stages and reward each one.
- Short sessions: a few minutes, several times a day, is better than one long session.
- End on success: finish with something your pet does well.
Choosing the right rewards
The reward must be something your pet values in that situation.
- Food: small, soft, tasty pieces, such as cooked chicken, cheese or high-value training treats. Use more valuable treats for difficult tasks or distracting places. Count them in the daily ration.
- Play: a short game of tug or a thrown toy for toy-motivated dogs.
- Praise and touch: pleasant for many pets, but often not enough on its own.
- Life rewards: going through a door, being let off the lead, sniffing a lamp post, greeting a friend.
- For cats: tiny pieces of meat or a lickable treat, and a wand toy for play-motivated cats.
Markers and clicker training
A marker is a signal that tells your pet "yes, that's it, a reward is coming". It can be:
- a clicker, a small device that makes a consistent click;
- a short word, such as "yes" or "good".
To start, "charge" the marker: click or say "yes", then immediately give a treat, ten or twenty times, until your pet looks expectantly at you after the sound. Then use it to mark the exact moment your pet does the right thing.
Ways to get the behaviour
- Luring: guide your pet into position with a treat, for example moving a treat over a dog's head so it sits. Fade the lure quickly so your pet does not depend on seeing food.
- Capturing: wait for your pet to do the behaviour naturally, such as lying down on its bed, and mark and reward it.
- Shaping: reward small steps towards a final behaviour, such as first looking at a mat, then stepping on it, then lying on it.
Adding a cue
Once your pet reliably performs the behaviour, add a word or hand signal just before it happens. Over time, your pet learns that the cue predicts the chance to earn a reward.
Reducing rewards over time
When a behaviour is reliable:
- reward most, but not every, correct response;
- vary the type of reward;
- use life rewards and praise more often;
- keep rewarding occasionally to maintain the behaviour, especially recall.
Dealing with unwanted behaviour without punishment
- Prevent it: manage the environment, for example by putting shoes away or using a baby gate.
- Teach an alternative: a dog that jumps up can learn to sit for greetings; a cat that scratches the sofa can learn to use a scratching post placed next to it.
- Remove the reward for the unwanted behaviour: for example, turn away from a dog that jumps up for attention.
- Meet the underlying need: more exercise, mental stimulation, chew items, or attention to fear and anxiety.
Avoid shouting, hitting, lead jerks, spray bottles, shaking cans and aversive collars. Remote-controlled electric shock collars are banned in England and Wales.
Training cats
Cats learn well with positive reinforcement, especially for:
- coming when called;
- getting into a carrier;
- accepting handling, nail clipping and medicines;
- using scratching posts;
- simple tricks such as "sit" or "high five".
Use very short sessions, high-value treats and a quiet environment.
Common mistakes
- rewarding too late;
- sessions that are too long or too difficult;
- repeating the cue many times ("sit, sit, sit");
- inconsistent rules between family members;
- accidentally rewarding unwanted behaviour, such as giving attention to barking;
- moving on to distracting places too soon.
Finding a trainer
Look for trainers and behaviourists who use reward-based methods and have recognised qualifications. In the UK, the Animal Behaviour and Training Council (ABTC) maintains registers of qualified trainers and behaviourists.
When to see a vet
Behaviour changes can be caused by pain or illness. Talk to your vet if your pet:
- suddenly becomes aggressive, fearful or withdrawn;
- has toileting problems despite training;
- seems to forget things it knew well, especially in old age;
- shows repetitive behaviours such as tail chasing or over-grooming.
Frequently asked questions
Won't my dog only obey when I have treats?
Not if you gradually reduce food rewards, use other rewards such as play and life rewards, and keep treats out of sight. Many well-trained dogs work for occasional rewards.
Can old dogs learn with positive reinforcement?
Yes. Dogs and cats can learn at any age. Older pets may need shorter sessions and gentler exercises.
Is clicker training better than using a word?
Both work. A clicker gives a very consistent, precise signal, which helps with complex behaviours. A word is always available. Choose what suits you.
Does positive training mean my dog has no rules?
No. Reward-based training is about teaching clear expectations and making the right choice easy and rewarding. It works well with boundaries and good management.
How often should we train?
Short sessions of 2–5 minutes, two to four times a day, plus rewarding good behaviour throughout daily life.
Paw4care editorial team
Sources
- Reward-based training for dogs, PDSA, accessed 2026-11-15.
- How to clicker train your dog, PDSA, accessed 2026-11-15.









