Positive Training: An Approach Built on Cooperation

Far from being a "soft fad", positive training is the scientifically validated method according to research in canine cognition. The American Veterinary Society of Animal Behavior (AVSAB) explicitly recommends using positive reinforcement exclusively and advises against coercive methods (e-collar, choke chain, alpha roll). Comparative studies show that a dog trained with positive methods is more obedient, less anxious and more creative than a dog trained with aversive ones. Here are the principles, techniques and practical examples.

The 4 quadrants of learning (theoretical framework)

Drawn from behaviourism (Skinner), 4 mechanisms change a behaviour:

• Positive reinforcement (R+): add something pleasant → the behaviour is repeated. ("You sit → treat").
• Negative reinforcement (R-): remove something unpleasant → the behaviour is repeated. ("You stand up → the head collar pressure stops").
• Positive punishment (P+): add something unpleasant → the behaviour stops. ("You jump up → shock from the e-collar").
• Negative punishment (P-): remove something pleasant → the behaviour stops. ("You jump up → I turn away and withdraw my attention").

Modern positive training uses only R+ and P- (the two ethically acceptable quadrants). It bans P+ (pain, fear) and limits R- (pressure).

Why R+ works better than punishment

Key studies to know:

• Hiby 2004 (Animal Welfare): dogs trained with rewards have significantly fewer behaviour problems than those trained with punishment.
• Cooper 2014 (PLoS ONE): e-collars cause more signs of stress and offer no advantage over R+.
• Vieira de Castro 2020 (PLoS ONE): dogs trained with aversive methods show more pessimistic behaviour in cognitive tests (a long-term effect).

Punishment works in the short term but damages the relationship, creates chronic stress and can turn an anxious dog into an aggressive one.

Timing: it all happens in under 1 second

A dog links a consequence to the action just before it only if the gap is less than 1-2 seconds. A reward given 10 seconds after a sit does not reinforce the sit.

• Use a verbal "marker" ("YES!") or a clicker to mark the exact moment of the behaviour you want.
• The marker means "well done, reward on its way!".
• The treat can follow within 3-5 seconds — the critical timing is the marker.

Teaching the clicker (charging → conditioning → use) takes 5-10 sessions of 5 minutes.

Rewards: not just treats

Rank rewards according to how hard the exercise is:

• Level 1 (simple learning at home): usual kibble, a kind word, a stroke.
• Level 2 (learning in a calm setting): dry treats, a short game.
• Level 3 (learning with distractions: park, street): high-value treats (chicken, cheese, dried liver).
• Level 4 (jackpot reward after a great exercise): several treats in a row + their favourite energetic game.

Play (tug, ball, find-it games) is often more motivating than food for some dogs.

Teaching a new cue: the shaping method

For a new skill (for example "paw"), break it down into micro-steps:

• Step 1: the dog looks at its paw → CLICK + treat.
• Step 2: the dog lifts its paw slightly → CLICK + treat.
• Step 3: it clearly lifts its paw off the ground → CLICK + treat.
• Step 4: it puts its paw in your hand → CLICK + treat.
• Step 5: name the action once it is learned ("paw") → CLICK + treat.

Short sessions (5-10 minutes), 2-3 times a day, always ending on a success.

Dealing with unwanted behaviour without punishment

Instead of punishing, aim to prevent, redirect, ignore or teach an alternative.

Example 1 — The dog jumps up at visitors:

• Prevent: keep the dog on a lead when guests arrive.
• Ignore: everyone turns away when the dog jumps up and ignores it completely (no eye contact, no words, no gestures).
• Reinforce the alternative: when the dog has all 4 paws on the floor → CLICK + treat + stroke.
• Teach "sit" for greetings: the dog learns that sitting gets attention.

Example 2 — The dog barks at other dogs on walks:

• Identify the threshold distance at which it starts to react.
• Work BELOW that threshold (far enough away for it to stay calm).
• As soon as it sees the other dog without reacting → CLICK + treat.
• Gradually reduce the distance over several weeks.
• This is the desensitisation and counter-conditioning technique (DSCC).

Tools to use and tools to ban

Use:

• Y-front harness: does not compress the windpipe and leaves the shoulders free to move.
• Long line of 3-10 metres for walks in open countryside.
• Clicker.
• Treat pouch on your belt.
• Snuffle mat and treat-dispensing toys for enrichment.

BAN (according to AVSAB and AVMA):

• Choke chain: dangerous pressure on the windpipe.
• Prong collar: pain, skin injuries.
• Electric / anti-bark collar: banned in several countries and condemned by veterinary societies.
• Water bottle / spray in the face: aversive, creates mistrust.
• Alpha roll (forcing the dog onto its back): the "dominance" theory has been completely disproved scientifically since 2009.

Dominance theory: a persistent myth

The idea of the "alpha pack leader" comes from flawed studies of captive wolves in the 1940s. Researcher David Mech, who had popularised the theory, officially retracted it in 1999 after observing wild wolves. Today, no serious ethologist defends this model to explain the dog-human relationship. Your dog is not trying to "dominate" you: it is trying to understand the rules and get what it wants.

Finding a good trainer

• Look for certifications: MFEC (French dog training instructor qualification), Pet Dog Trainer training, IMPACT-COE certifications, the "Pat Miller" or "Karen Pryor" methods.
• Ask to watch a session before you sign up.
• Steer clear of trainers who talk about "dominance", "pack" or "leader", or who use e-collars, choke chains or intimidation techniques.
• A good trainer always explains the "why" behind each technique.
• For complex behaviour problems: see a veterinary behaviourist (a specific university qualification).

Conclusion

Positive training is not a "nice" option reserved for small dogs: it is the scientifically validated method for ALL dogs, including large "difficult" dogs and working dogs. It builds trust, develops cooperation, and produces dogs that are both obedient and happy. Aversive methods belong to the past — science and ethics are unambiguous on the subject.