A learning process in which a behaviour increases because performing it produces something the animal finds rewarding. The “positive” refers to addition (something is added), not to the behaviour being good or the animal being well-behaved.

A dog sitting on cue and receiving a piece of chicken, a horse standing quietly at the mounting block and receiving a wither scratch, a parrot stepping onto a hand and receiving a favourite seed, are all examples of positive reinforcement. The behaviour produces something the animal values, and the behaviour becomes more likely in future as a result.

Positive reinforcement is widely used in clicker training, in zoo husbandry training where animals are taught to voluntarily cooperate with medical procedures, in service dog training, and increasingly in horse work as the evidence base for its effectiveness and welfare benefits has grown. It is the dominant reinforcement mode in modern applied animal behaviour practice.

A common concern about positive reinforcement, particularly with food, is that it produces pushy or dependent animals. The research evidence does not support this concern when positive reinforcement is applied with appropriate protocols. Pushiness is typically the result of inconsistent reinforcement (the animal has learned that pushy behaviour sometimes works), not of positive reinforcement itself.

The effectiveness of positive reinforcement depends substantially on timing (the reinforcer must arrive promptly after the desired behaviour), value (the reinforcer must matter to the individual animal in that moment), and consistency (the contingency between behaviour and reinforcer must be reliable enough for the animal to learn the association).