Leena barks, lunges and generally loses her mind when she sees other dogs on a walk. Her owner and I are walking in the park so I can teach the owner how to change Leena’s behavior. When we see another dog at a distance, Leena freezes and her head comes up, mouth closed. She’s about to fall off the proverbial cliff. I put a handful of chopped chicken in front of her nose and start feeding her rapidly. She focuses on eating the chicken and does not bark or lunge toward the other dog.
The owner watches this and says, “Aren’t you just distracting her, though?” Am I? Is it bad if I am? This example highlights one of the many points of confusion when it comes to using food to change behavior. You’ve probably heard people refer to using food to ‘bribe’ dogs. Other common words are ‘reward’ and ‘reinforcer’. These terms are sometimes used interchangeably, but they actually mean quite different things. What’s the difference and how do these terms come into play in the real world?
Reinforcer vs. Reward
Let’s start first with the most technical term – ‘reinforcer’. By definition, a behavior increases if it is reinforced. Therefore, a reinforcer is anything that increases a behavior. Typically, the term ‘reinforcer’ is used in the context of positive reinforcement where something desired (the reinforcer) is being given to the dog to increase or strengthen a behavior. (For a more detailed discussion of reinforcement, I suggest reviewing my blog post on reinforcement and punishment.) Reinforcers can be used to teach new behaviors or maintain existing ones. They can also be used to modify or strengthen a behavior that’s already known – such as increasing the length of a stay or getting a straighter sit for obedience. Examples of potential reinforcers are food, toys, and access to anything the dog enjoys such as attention, going for walks or smelling a mailbox.
Many people use the term reinforcer and reward interchangeably; however, they are not quite the same thing. A reward is something that is given in an attempt to increase behavior. Rewards can fail in that they don’t actually increase behavior. For example, if someone offered to send me to Paris as a reward for running a mile, I almost certainly wouldn’t do it. I do not have a lot of interest in visiting Paris. I’m also not a runner so running a mile would involve quite a lot of effort for me. In this case, a trip to Paris is a reward, but not a reinforcer – because it doesn’t increase my behavior of ‘running’. However, if a trip to Alaska was offered as a reward for running a mile, I would probably do it. In this case, a trip to Alaska is a reward (an attempt to increase behavior) and a reinforcer (it actually increases behavior).
Side note – the Paris example also highlights the importance of motivation in training as well as the role of effort. The use of a high value reward is more likely to increase behavior. What is high value varies from dog to dog (and person to person). Paris is not a high value reward for me. Alaska is. It might be different for you. You may have noticed in the previous paragraph, I said “examples of potential reinforcers…” This is because not all rewards are reinforcing to all dogs. My beagle mix, Darwin, doesn’t care about toys and absolutely would not work to earn a toy. My Australian Shepherd, Finn, on the other hand, would possibly sell his soul for a game of fetch or tug.
Effort matters too. I don’t run, but I do bike. If someone offered to send me to Paris for a 10-mile bike ride I would be much more likely to do it because I enjoy biking anyway and biking 10 miles is not difficult for me. This could be a very in-depth discussion, but I am getting sidetracked away from my main point, so let’s get back to that!
One common complaint that comes from people that don’t want to use food in training is that they don’t want to bribe their dog. Is that what’s really going on? The Cambridge English Dictionary defines ‘bribe’ as “the act of giving someone money or something else of value, often illegally, to persuade that person to do something you want”. In the case of the bribe, the valued item is offered before the desired behavior, in an attempt to convince or persuade the receiver to do the behavior.
Many pet dog owners use food this way. However, when used properly, this is not what happens with reinforcement. This is very important! Proper reinforcement happens after the performance of a behavior. Food should not be held out as an enticement to convince the animal to do something for you. Why not? Because this will make it very difficult to get the animal to do the behavior without proof of reinforcement ahead of time. If you withhold the reinforcer until you get the behavior you want, and then offer it to the dog, it will go a long way toward avoiding the issue of being dependent on your reinforcer.
One exception to this rule is the technique of luring. With luring, you let the dog smell the food in your hand (it does not need to be visible). You then use the food as a guide to get the dog to do a behavior (such as a sit) by following the food hand. We could argue the semantics here (and you are welcome to speak up if you disagree with me!), but I would consider this communication rather than bribery. In this case, you are using the food to show the dog what you would like her to do. Once she figures it out (for example, sits), then the best thing to do is remove the lure as quickly as possible and only present the food (or another reinforcer) after the dog does the behavior.
There is another context where I often use food. I frequently want to teach a dog to associate a particular trigger (such as another dog or a stranger) with something good (usually food). Let’s go back to Leena. In this case, I pair the appearance of the other dog with food. This is an example of classical conditioning (specifically classical counter conditioning) where I am associating one stimulus (dog) with another (food) in order to change Leena’s emotional response. This is not reinforcement! I am not asking her for any particular behavior – I am simply teaching her that Other Dogs = Awesome Food!
This brings us back to the question, “Aren’t we just distracting the dog?” If Leena has already seen the other dog, then the answer is no. I am not distracting her. I am teaching her to associate other dogs with high value food. This is necessary because it will help bring her out of that state of high anxiety or excitement. It is difficult (or impossible) to teach a preferred, calmer behavior (such as heel or eye contact) when the dog is in a state of high emotional arousal.
If Leena has not already seen the other dog, then yes, the food would be a distraction. Is that a bad thing? Not in my book! The food is one tool for avoiding (or reducing) the lunging and barking which is good because we don’t want her practicing that behavior! Distraction can be very valuable in certain contexts. However, if it is the only method used you are much less likely to see dramatic behavior change in the long run because the dog will still react if they see the other dog (or whatever the trigger is) before you have a chance to distract them. For this reason, the best approach is a combination of the following:
- Distraction and/or avoidance when the situation is likely to be too hard for the dog
- Classical counter conditioning when the trigger is mild enough that the dog doesn’t become overly aroused (that is, barking, lunging, refusing to eat, etc.)
- Reinforcement of desired behaviors (like heel and eye contact) when the dog is emotionally calm enough to focus.
Here is a quick summary:
- Reinforcer – increases behavior
- Reward – attempt to increase (or show appreciation for) a behavior. May or may not actually increase behavior
- Bribe – presented before the behavior is completed in order to persuade the animal to do the behavior
- Lure – used to show a dog a new behavior by following a treat. Should be faded quickly once the dog is doing the behavior.
- Classical conditioning – creates an emotional response where one thing predicts the presence of something else (for example, dogs predict yummy food). Very helpful for changing a dog’s emotional response and decreasing his arousal level so that he is more receptive to learning a replacement behavior
There you have it! I hope this post has helped you to sort out the differences between reward, reinforcer, classical conditioning, bribery, and distraction. If you already had a good handle on this information, hopefully, it will serve as a useful resource for your clients. The better we understand the different terms – and the best way to apply them – the better we will be at improving our dogs’ emotional wellbeing and behavior!
Looking for more? Dog professionals can check out our professional services page. Dog owners in Saratoga County, NY can find more information on our local services here. What other questions do you have about behavior and learning terms? Post your answers in the comments section!