A common question when it comes to animal training is:
Why are we using food to train?
Do we have to use treats?
Can’t I just use praise or petting?
Let’s dive into why we use food in training and why it can be so beneficial to the learning process. Doing so requires knowing a little about the science behind it and a reminder that only the learner gets to decide what is actually reinforcing to him.
Types of reinforcers*
There are several different categories of reinforcers, but let’s just focus on primary and secondary reinforcers for now. Primary reinforcers are those things that are necessary for survival: food, water, shelter, etc. Because they’re necessary for survival, all individuals find these reinforcing in most situations. Secondary reinforcers are things that have been paired with primary reinforcers so that they too become reinforcing: toys (paired with play and fun), petting (paired with physical contact), etc. Because they must be learned, they are not inherently reinforcing to all individuals.
As mentioned, food is a primary reinforcer because it’s necessary for survival. That’s why it’s so easy to use food in training; it’s [almost] always desirable! Whereas secondary reinforcers, which can be powerful training tools, need to be paired with a primary reinforcer first and in a way that makes them just as powerful. Essentially, we have to do extra training in order to make those things as successful. This is why toys may be reinforcing for one dog but not for another, whereas food is generally reinforcing for both. In short, primary reinforcers are more likely to actually be reinforcing without additional training and among those, food is usually the easiest to dispense.
*Side note: there’s a problem with how I’m talking about reinforcers in sweeping generalizations in this section which goes back to only the learner decides what’s reinforcing. We have to actually observe the behavior to see if it’s increasing or decreasing to deem something as reinforcing in that context. It’s not enough to just apply a reinforcer and assume that it’s going to work in the way that we intended; observation is vital. Additionally, reinforcers are not always desirable (yeah, behavior is weird sometimes) and so these sweeping generalizations can get dicey. In short, know that there’s a lot more to the story above and that this is just meant to dip your toe into this topic as a pet parent.
When speaking with pet owners, I find that the concerns about using food vs. another type of reinforcer have less to do with the actual science and “why”, and more to do with one of the following concerns.
“I’m worried about them gaining too much weight.”
I often hear something along the lines of, “He’s going to weigh 300 lbs at the end of this!” while speaking with clients about their pet’s behavior modification plan. While it’s said as a joke, it really is a way of voicing concern about weight gain. Others come right out and tell me that they’re worried about their pet gaining weight because of training. It can be a valid concern, especially for certain breeds.
If we’re working with a young, growing pet or a high-energy individual who hasn’t had weight concerns before, I tell my clients that while it’s something to keep in mind, we don’t necessarily have to be immediately concerned. Let’s start the behavior modification process and if we see some weight gain then let’s adapt accordingly. If we start to see problems or if it’s an individual or breed who is prone to weight gain, then we have some options:
- Use smaller treats (I break small training treats in halves or quarters even for Oso)
- Experiment with fruits and veggies for treats
- Set aside some of their meals to use for training
- If it’s a treat-heavy day then give them a little less during meal-time
“I’m worried they’ll get an upset stomach.”
There are many people who’ve experienced their pet having an upset stomach due to the type of treat or having too many treats. I often hear of pets throwing up or having diarrhea after hour-long training classes. Some pups will keep eating even if they’ve surpassed their limit! If this is something you’ve experienced, keep in mind that training sessions at home should never be that long.
As trainers, we have to compromise in providing longer services (aka training sessions) because it logistically works better; we wouldn’t recommend actually training that long (which is why there needs to be so much down time where the pet isn’t working during those services). I generally start with training for 2-3 minutes, then take a break, then train for another few minutes. Many pets are done after 10-15 minutes of this repeating cycle. Some pets (*cough cough* puppies) need less training time than that whereas seasoned learners can go for longer. Short bursts of training can be quite effective. And, short sessions means you’re not plying them with food for an hour at a time, causing upset stomachs.
The other factor behind this concern is for pets who have naturally sensitive stomachs. In that case, we should be speaking with the vet about what foods they can have and limiting the ingredients to what their stomach does well with. Sometimes we have to get creative when it comes to certain diets; a professional can help with that since it’s more specific to your individual pet’s needs.
“I don’t want them to become dependent on food.”
Cheeky response warning: I hate to break it to you, but if they’re alive then they’re dependent on food. We all need food to survive and that’s where the dependence comes from, not from using it in training.
What I think people really mean when saying this is…
“Do I have to use food for training forever?”
It depends. It depends on if you’re talking about one specific behavior or when teaching new skills, your pet, and a few other things. If you’re talking about always using food while teaching new skills, I would say the answer is, “Sure, because it’s easier to do so.” You could absolutely beef up your other reinforcers so that they’re as effective as food is at teaching new behaviors, but the “work smarter not harder” response is to continue using food for teaching new skills.
If we’re talking about one specific behavior, then my answer is still, “It depends.” Oso has a few behaviors that I never treat him for; I use “real-life reinforcers” to maintain the behavior. For instance, he knows an “up” cue for jumping on the furniture or jumping into our laps. The reinforcement for doing those behaviors is proximity to us (which his behavior has said is reinforcing) and being on the comfy furniture. However, there are behaviors that I will always treat him for because they’re a matter of safety and they need to be super reliant, like coming when called. Real-life reinforcers aren’t enough to maintain that specific behavior for him in all situations.
To break down that “it depends” into a more concrete answer: It depends on whether other things– like petting, praise, toys, etc.– are enough to maintain the behavior to the level you want/need it to be. If yes, great. You can maintain the behavior with other reinforcers. If no, then continue using food.
“It’s difficult to have food on me all the time.”
If you’re one of those people, like me, who doesn’t enjoy having treats in their pockets then yes, this is true. (More power to those people who do it, though!) To get around this, I have treat jars set up around my house for easy training and recommend that folks keep a stocked treat pouch attached to their leash for easier access.
“Food doesn’t always work.”
Even though it’s a primary reinforcer, there are reasons why food would not be reinforcing in certain situations. In more extreme cases, I’ve seen where punishment related to food has essentially made food itself scary. But the more common reasons are that we’re asking them to do something that’s not worth the food or that they’re not able to learn at that moment.
Let’s use a human example to explore when food isn’t worth it. Say you’re doing one of your favorite activities and someone approaches you and tells you that you need to stop what you’re doing and do your taxes instead. Fat chance, right? Maybe they offer your $5 to switch tasks. Thanks but no thanks. Now let’s say they offer you $50,000 to stop what you’re doing and do your taxes right now instead. I don’t know about you, but I would jump on that opportunity. If you were given $50,000 for switching from your favorite activity to doing your taxes enough times then you would readily make the switch because you have a reinforcement history for doing so with a reinforcer that’s worth it.
Now, let’s apply that to an example that I mentioned earlier: I always use food as a reinforcer for Oso’s recalls. Oso is outside doing one of his favorite activities of hunting for rodents. I need him to come inside so that I can leave; he’ll be left alone with something to do but it’s definitely way less fun than hunting for rodents. I could offer him the Oso-equivalent of $5 for coming inside: a kibble. Or I could offer him the Oso-equivalent of $100 for coming inside: treats he loves. A kibble just isn’t worth it; it needs to be better than that. Oso now has a strong enough reinforcement history that I don’t necessarily need to give him the treat but because I want to keep it as a super reinforcing behavior I will continue to do so.
The other reason why food might not work in that moment is because of stress, fear, or anxiety. Food isn’t reinforcing when you’re protecting yourself from a perceived threat. We’re just focused on living to eat another day at that moment. I wrote an article dedicated to this here, titled “Does Your Pet Have Mountain Lion Brain?”
- If you’re hesitant to use food in training, which of the above reasons most resonates with you?
- Take the time to read through and sit with that point. What concerns do you still have?
- Speak with your behavior consultant or trainer about those concerns; they can help!
- Get to training and make your own observations. What reinforcers build super strong behaviors in your pet?