What the Science Actually Says About Dog Training

By The Bark Science Research Team · September 17, 2026 · 5 min read

Why Training Methods Matter More Than You Might Think

Dog training is one of those topics where strong opinions are common and rigorous evidence is harder to come by. Over the past two decades, however, a growing body of peer-reviewed research has begun to give us a clearer picture — and the findings consistently point in the same direction. How you train your dog has consequences that extend well beyond whether they sit on command.

What Reward-Based Training Actually Involves

Reward-based training — sometimes called positive reinforcement training — works by marking and reinforcing behaviours the dog offers that you want to see again. The reward can be food, play, praise, or access to something the dog finds enjoyable. The dog learns that certain actions produce good outcomes, and those actions become more frequent over time.

Crucially, reward-based training is not simply about handing out treats. It involves timing, consistency, and clear communication. When done well, it teaches a dog to actively participate in learning rather than simply to avoid punishment.

If you're not sure where your own training approach currently sits, taking a short quiz can help you identify which methods you're already using and where there might be room to develop.

What Aversive Methods Involve — and What the Research Finds

Aversive training methods apply something unpleasant — a shock, a sharp sound, a physical correction — to discourage unwanted behaviour. The logic seems straightforward: the dog experiences something uncomfortable when it does the wrong thing, so it stops doing that thing.

The research, however, suggests the reality is more complicated. Studies examining the use of aversive tools and methods have found associations with elevated stress indicators in dogs during and after training sessions. Dogs trained with aversive methods have been observed to display more stress-related behaviours, including lowered body posture, yawning, lip-licking, and avoidance, compared to dogs trained with reward-based approaches.

A review of the available literature on aversive training methods concluded that their use is associated with a range of negative outcomes for dog welfare, including increased fear and anxiety, as well as a potential deterioration in the dog-owner relationship. The review also noted that the existing evidence does not support the idea that aversive methods are more effective than reward-based alternatives at achieving reliable behavioural outcomes.

The Question of Effectiveness

One common assumption is that aversive tools might be "harsh but effective" — a trade-off between welfare and results. The research does not support this framing.

Studies comparing remote electronic collar training directly with reward-based training found that both approaches could produce behavioural change, but that reward-based methods achieved this without the welfare costs associated with collar use. Dogs in the aversive training groups showed more indicators of stress, without demonstrating superior learning or compliance.

This is an important point. The justification for using aversive methods often centres on their supposed effectiveness for difficult cases or highly distractible dogs. But when that effectiveness is put to the test under controlled conditions, the advantage largely disappears — while the costs to the dog's emotional state remain.

Why Stress Matters to Learning

Understanding why aversive methods undermine training outcomes requires a brief look at how stress affects cognition. When an animal is in a heightened stress state, its capacity for flexible, associative learning is reduced. A dog that is anxious or fearful is focused on the threat, not on working out what behaviour is being requested.

Reward-based training, by contrast, tends to produce a dog that is engaged and motivated — what researchers sometimes describe as an "optimistic" emotional state. Dogs in this state are more likely to offer behaviours, try new things, and generalise what they have learned to new contexts. The emotional environment of training is not incidental to the outcome; it is part of the mechanism.

The Relationship Between Dog and Owner

Research in this area has also examined what training methods do to the broader relationship between a dog and the person training them. Dogs trained with aversive methods have been found to show less engagement with their owners and more signs of anxiety in their presence. Reward-based training, by contrast, appears to support a more secure and cooperative bond.

This matters practically. A dog that trusts its owner is easier to manage in novel or stressful situations. That trust is built through thousands of small interactions, and training is one of the most concentrated of those interactions. Thinking carefully about how you train is therefore not just a philosophical choice — it is an investment in the quality of your relationship over the dog's lifetime.

What This Means in Practice

The scientific evidence on training methods is not yet complete — no field ever is. But the current consensus is clear enough to draw some reasonable conclusions. Reward-based methods are at least as effective as aversive ones, and they achieve their results without the stress, fear, and welfare costs that aversive approaches carry.

For anyone working with a dog — whether a new puppy or an older dog with established habits — this evidence provides a sound basis for choosing methods that work with a dog's learning capacities rather than against their emotional wellbeing.

References

Vieira de Castro, A. C., Fuchs, D., Morello, G. M., Pastur, S., de Sousa, L., & Olsson, I. A. S. (2020). Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. PLOS ONE, 15(12), e0225023.

Cooper, J. J., Cracknell, N., Hardiman, J., Wright, H., & Mills, D. (2014). The welfare consequences and efficacy of training pet dogs with remote electronic training collars in comparison to reward based training. PLOS ONE, 9(9), e102722.

Ziv, G. (2017). The effects of using aversive training methods in dogs—A review. Journal of Veterinary Behavior, 19, 50–60.