EnrichCove Evidence Card: Force-Free vs. Aversive Dog Training (Printable)

The main research on reward-based versus aversive dog training, one row per source, with each finding written to match the original paper. Print it (Ctrl+P or Cmd+P) or bookmark it. It goes with our article Force-Free vs. Traditional Dog Training: What the Science Actually Says.

Source What it found
Hiby EF, Rooney NJ, Bradshaw JWS (2004). Dog training methods: their use, effectiveness and interaction with behaviour and welfare. Animal Welfare 13(1):63-69.
Questionnaire survey of 364 dog owners (authors at the University of Bristol, UK); correlational, owner-reported 		Owner-rated obedience correlated with the number of tasks trained with rewards; the number of reported problem behaviors correlated with punishment use, not reward use.
Cooper JJ, 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.
Field trial of 63 pet dogs referred for recall problems, 3 groups of 21 (e-collar; same trainers without e-collar; reward-focused APDT trainers); video, cortisol, owner reports 		E-collar dogs spent more time tense, yawned more and interacted less with their surroundings than the reward-trained group; owner-reported improvement did not differ between groups.
Ziv G (2017). The effects of using aversive training methods in dogs: A review. Journal of Veterinary Behavior 19:50-60.
Literature review of 17 studies (surveys, observational studies and interventions) 		Aversive methods can jeopardize dogs' physical and mental health; no evidence showed positive punishment outperforms positive reinforcement, and some evidence suggested the opposite.
Vieira de Castro AC, Fuchs D, Morello GM, Pastur S, de Sousa L, Olsson IAS (2020). Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. PLoS ONE 15(12):e0225023.
Quasi-experimental (non-randomized) study of 92 companion dogs from 7 training schools: 3 reward-based (Reward, n=42), 2 using low proportions of aversive methods (Mixed, n=22), 2 using high proportions (Aversive, n=28); video, saliva cortisol (subset of dogs), cognitive bias task 		Aversive-group dogs showed more stress behaviors, larger post-training cortisol rises and more "pessimistic" responses than Reward-group dogs; Mixed-group dogs differed only in stress behaviors.
China L, Mills DS, Cooper JJ (2020). Efficacy of dog training with and without remote electronic collars vs. a focus on positive reinforcement. Frontiers in Veterinary Science 7:508.
Re-analysis of the same 63-dog field trial as Cooper 2014 (3 groups of 21, up to 150 minutes over 5 days); objective command response and latency 		The positive-reinforcement group responded better to "Sit" and "Come" after one cue; the authors concluded there is no evidence e-collar training is necessary.
American Veterinary Society of Animal Behavior (2021). Position Statement on Humane Dog Training. AVSAB.
Professional position statement summarizing published research on dog training methods 		AVSAB recommends reward-based methods for all dog training and behavior modification, and says aversive methods, including shock, prong and choke collars, should not be used.
Mech LD (2008). Whatever happened to the term alpha wolf? International Wolf 18(4), Winter 2008.
Popular-science article by the wolf biologist who popularized the term, summarizing field research on wolf packs (not a dog study) 		Mech wrote that most wolf packs are parent-led families; the alpha idea came from studies of wolves from different zoos housed together.

Notes: Cooper et al. 2014 and China et al. 2020 report the same 63-dog trial. Hiby et al. 2004 is a correlational owner survey, and Vieira de Castro et al. 2020 is quasi-experimental, so neither shows cause and effect on its own.