Beyond the Operant: What We Mean
Learner centered training is not literally beyond operant conditioning, but it does take the concept and apply it in innovate, interdisciplinary ways that transform our lives with our pets.
We start with the belief that agency is an antecedent. It is not the result of our training and it is not the same as choice. If we use Bandura’s 1977 definition, agency is an individual’s belief in their ability to exert control over their own independent functioning and external environment. Notably, this doesn’t require the individual to have control or even to have choice necessarily, but to believe they can exert independently-generated control over their environment.
It is a fact in human psychology that agency increases distress tolerance - the research verifying this effect is plentiful. People who opt into stressful situations persist longer than those who are coerced or forced into the same situations. Further, people who identify values-based motivation for sticking with an aversive situation or task stick with it even longer than they do an arbitrary task. For instance, someone who is voluntarily participating in exposure therapy so they can go on a trip with a beloved family member, for example, will persist much farther into the aversive exposure than someone who is told to keep their hand in ice water for as long as they can for $5 or some other arbitrary, extrinsic reinforcer. This is why a reactive dog who has been taught gestural communication to indicate their directional preferences on a walk can handle triggers significantly better than their handler(s) would previously have believed.
Another key factor in learner centered training is that it centers communication and feedback directly from our animal learners, as seen in the reactive dog example above. Learner centered training builds on concepts of communication and values-alignment that come from Relational Frame Theory. Anyone who has spent time with a baby has seen or experienced the frantic search for why the baby is screaming - trying food, milk, cuddles, putting them down for a nap. Babies don’t have a way to ask for anything specific, so frustration builds as they do not get what they want or need and are not being heard, or understood. There are two overlapping contingencies, one about wanting a thing, and the other about the communication. We see the same thing even more clearly when children are just learning to talk. They tell us verbally they want something, but heck if we can understand what their cute gibberish means, and we see their frustration build as time goes on while we try to figure it out – frustration generates from not being understood, which compounds the unmet, original need.
Relational Frame Theory postulates that, as we learn communication, the people around us encourage us – when a kid sees a dog and the parents go, “doggy!” and the kid says, “goggy!” and the parents confirm, “yes, doggy!” they are providing social reinforcement for the child’s use of language. As the child learns what words and - later - what ideas and actions, are reinforced by parents - or, later - peers, they determine how they’re supposed to navigate the world around them. Communication, shared beliefs, and values-aligned actions can all be forms of reinforcement and can act to alter motivation.
So, with our dogs, when we teach them a mand - or a way to request what they want - we are opening the door for them to choose to use that mand any time, in any context, at will. A dog who knows to boop their handler’s knee for a cookie can do that in the kitchen, on the street, or at the vet. Once dogs can mand, they can exert independently-generated control over their environment - increasing overall agency. We are also answering their mands, which takes us from a simple demand to a two-way conversation, or what’s called an intraverbal. We all like to feel heard, even if the answer is no - the frustration of being unheard or misunderstood goes away, and we’re left with the unmet need underlying the request, which may or may not have big feelings attached. Either way, the additional frustration related to communication diminishes when we receive an answer - any answer - which decreases overall frustration.
That decrease in frustration is related to a change in motivation. Motivation for behavior in the moment goes from “get the thing” to “ask for the thing.” Even if the answer is no, we have still successfully communicated (using intraverbals) and that clarity is, in itself, reinforcing. The ability to request things clearly, and then receive them, is proxy agency - basically, we, the handlers, are facilitating the learner’s impact on the environment. As agency increases in dogs’ overall lives, they can tolerate distress better and, in a super neat thing to see, they can generate their own communicative gestures. My dog currently has more management strategies generated by her than by me!
As a quick comparison, a dog who jumps up at guests at the front door is often taught a go to mat behavior. We could instead teach three options
An option that gets them pets; say, sitting calmly next to their human.
An option that indicates they do not want to have any interaction; say, going to another room.
An option for being left alone but that preserves their ability to watch and evaluate; say, go to mat. They may then decide to return to the human’s side to request pets later, once they’ve had some time to observe.
Teaching three options, any of which is acceptable in the circumstance, is a shift from offering the dog a choice of two things and letting them choose. When applied across contexts, the learner centered approach empowers animals to request an infinite number of things. Rather than the paternalistic two-option approach – though kind and very well-intentioned, but with us fully in control - our learners can weigh in on both the decisions and help identify available options. Therefore, they tolerate low-choice moments better and we can observe patterns in their choices that help us better meet their needs in low-choice moments. When we have a good idea what they would choose, we can make more informed choices about alternatives when the most preferred choice isn’t available.
Something as simple as manding for a choice of toy can be the foundation for more meaningful mands, but we also might learn the dog really prefers fetch over tug, or only likes rubber toys for outside play, or any number of interesting preferences. We could also see them request fetch after days with little activity, maybe to burn off excess energy, or a chew toy when they’ve been running around all day, or vice versa! Often they surprise us, and they can meet their own needs better when they can tell us what they need. We can also gather data based on their choices, like whether they want proximity or distance from triggers. Barky lungey behavior leaves us guessing and taking over control in the situation; a mand maintains calm behavior while equally benefitting the dog.
When we say learner centered training is beyond the operant, we mean it goes far beyond the applications of operant conditioning to date. We focus on reinforcing intraverbals and values-alignment rather than very specific behaviors, to create a more meaningful experience for our animals while still meeting our needs as their guardians/owners.
