Aversive-based training focuses on punishment and negative reinforcement such as use of shock, pinch and choke collars, leash jerks, physically dominating, striking or yelling at the dog. In contrast, reward-based training focuses on positive reinforcement; for example, giving the dog food treats. There are concerns about poor animal welfare outcomes from aversive-based training compared to reward-based training. However, few studies compare dog welfare during and after different types of training.
This study investigated the effects of different training methods on the welfare of companion dogs. A total of 92 dogs were recruited from 7 training schools in Porto, Portugal. Dogs at 2 schools received 75 to 84% intended aversive training (Group Aversive, n=28 dogs). Dogs at 3 schools received only reward-based training (Group Reward, n=42). Dogs at 2 schools received less than 37% aversive training (Group Mixed, n=22). During and after training, dog behaviour was evaluated including stress-related behaviours (e.g., lip licking, yawning) and overall behavioural state (e.g., tense, low, relaxed, excited). Saliva samples were collected to measure the stress hormone cortisol. In addition, 73 of the dogs completed a cognitive bias test where latency to reach stimuli is thought to reflect affective state (emotion). Statistical analyses controlled for potential confounders including owner gender, children in household and dog age.
Group Aversive showed more frequent stress behaviours and low and tense behavioural states than Group Reward. Average post-training increase in salivary cortisol was higher in Group Aversive than in Group Reward. In the cognitive bias test, Group Aversive demonstrated longer latencies for all the stimuli suggesting they regarded the food reward as less probable possibly due to a less positive affective state. While this was not a randomised control trial that could establish causality, the study is the first to systematically demonstrate that companion dogs trained with aversive-based methods experience poorer welfare compared to dogs who receive reward-based training.
de Castro ACV, Fuchs D, Morello GM (2020) Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. PLOS One 15(12), e0225023.