OPINION
Suggestions to Rescue Reward-Training from Aversive Techniques
By Dr . Ian Dunbar , APDT Founder
I founded the APDT as an open educational forum to create a large membership , so that dog trainers had a concerted and powerful voice to control their own destiny and be able to thwart challenges from other professions , e . g ., the veterinary profession , or government agencies . Now , I think the APDT and CCPDT desperately need to rescue reward-training from its memberships ’ creedal worship of a laboratory learning theory that has tainted the training / teaching techniques and motivation of all animals , humans included , for more than a century .
Many reward-training techniques have gone down a theoretical rabbit hole , taking practical application and effectiveness along for the downhill ride . Thus , allowing leash-corrections and shock to creep back into public view ( social media ). Aversive techniques had been virtually eliminated in the 80s and 90s because they simply could not compete with the reliability of off-leash lure / reward-training for teaching reliable , verbally cued responses , that also provided non-aversive solutions for misbehavior and non-compliance . When I studied learning theory at Berkeley , I thought , “ Well , this won ’ t work too well for training dogs off-leash .” Focusing on a single behavior at a time and with only four ways to change behavior is not that sophisticated and hardly realistic . Behavior comes in lengthy , rapid-fire sequences , changing qualitatively several times a second and so feedback ( differing consequences ) must reflect this . Using the same-oldsame-old quantum food reward over and over won ’ t suffice ; reinforcement must be analogue , which is best accomplished verbally .
Ian ’ s Definition of Training Teaching dogs to the eventual criterion of responding promptly , reliably , and happily to verbal instruction , in any scenario , including when offleash , at a distance , and distracted , and without the continued need for any training aid whatsoever , especially including , food lures , food rewards , hand-contact , leashes , collars ( including metal and shock ), halters , and harnesses . You ’ ll notice that all three of my woolly adverbs may be objectively quantified . When Response-Reliability percentages are tested before and after a single training session ( Test-Train-Test ), the difference in results offers an index of improvement , and speed of improvement . Moreover , I routinely train dogs off-leash in safe surroundings , so that I can continually evaluate the dog ’ s consent to join me in the training game . The profession has changed so much over the past four decades . First the abrupt , 1910 transition from off-leash training with verbal instruction and guidance to formalized on-leash obedience , primarily using leash ‘ corrections .’ Second , the 1982 revolution back to off-leash , non-aversive , lure / reward , puppy socialization and training group classes to prevent
predictable , adolescent-onset fear of unfamiliar people and dog-dog reactivity , and to teach dogs to respond reliably to verbal cues when off-leash .
When I first introduced ‘ learning theory ’ to dog training in 1987 , I cautioned trainers should not blindly follow the reinforcement schedules or punishment techniques , but to understand the principles and relative pros and cons . I suggested trainers be Border Collies and not sheep . The 100-year-old reinforcement schedules are fraught with problems and are relatively ineffective ( CC , FI , and VI ). Some schedules are motivating and maintain responding ( VI and VR ), but trainers can neither create nor apply them . The principle we should have learned is that rewards must always be administered unpredictably .
Aside from sabotaging reinforcement , the considerable harm these old studies inflicted on animal training was to give ‘ aversive stimuli ’ the same star billing as ‘ rewards ’. They made a ‘ poster child ’ of aversive stimuli , creating a false Reward vs . Punishment binary . My view : Aversive Stimuli should never have deserved to be on the same stage as Rewards . Not even backstage , not even the understudy ’ s dressing room , or the dressing room toilet ! It must be obvious to everyone that aversive stimuli seldom work to inhibit unwanted behavior outside of captive populations ( caged , leashed , or restrained animals , children , spouses , students , employees , and prisoners ),
The APDT Chronicle of the Dog | Spring 2024 15