A few weeks ago, I made this post for our Instagram page showing an example of contrafreeloading in my bird room. In the video, the birds had eaten every crumb out of a homemade food puzzle but left quite a bit of food in a bowl sitting nearby.
For those of you who are unfamiliar with the term “contrafreeloading”, it refers to a phenomenon where animals, when given a choice between freely available food and food they have to work to obtain, choose to work for their food. Sounds intriguing, right? Keep reading, because it gets even cooler.
And for those of you who are already familiar with the term, I’m going to ask you to keep reading anyway. Because here’s the thing: we humans have a tendency to take a complex scientific concept and oversimplify it, or take it to unfounded extremes–and that is exactly what has happened to the concept of contrafreeloading. So let’s go on a little journey together to dive more deeply into what contrafreeloading is all about.
The Debate Over Animal Effort
Before we get into the details of what contrafreeloading is, it’s probably helpful to first describe the polarized debate over asking animals to work for their food, because this debate gives us important information about how contrafreeloading is being applied, misapplied, or ignored in the animal welfare community.
On one side, the argument is that making animals work for their food is cruel, because we’re controlling access to a resource that should be freely available to them. Proponents of this argument talk about the cascading harm of food insecurity and disempowerment for the welfare and well-being of these animals. This argument comes from a good place, in that we do owe it to domesticated and captive-bred animals, who have lost the ability to control their own lives and fend for themselves, to give them as much care and empowerment as possible. However, this argument stems from either being unaware of the concept of contrafreeloading, or from misunderstanding research papers like this one and mistakenly believing that “contrafreeloading has been debunked.” Don’t worry: we’re going to come back to this paper, because it’s a very cool example of how complex contrafreeloading can be!
On the other side, people often use the idea of contrafreeloading to justify a NILIF (Nothing In Life Is Free) philosophy, in which they believe that animals must earn everything they get. Proponents of this argument believe that freely available food will spoil animals, or that it is the root of most behavior problems. Within this camp, there is a huge push to get rid of food bowls, because eating food out of bowls is bad for our pets. This argument also comes from a good place, because it is also concerned with empowerment, and the belief is that teaching animals how to work for their food is empowering. However, this argument also demonstrates an incomplete understanding of contrafreeloading.
In reality, even though these two seemingly opposing arguments seem incompatible, there’s actually a whole lot about both of them that is really good and worth keeping. And you know what? They’re not actually mutually incompatible at all! What this debate needs are some dialectics–which means digging deeper to understand how two or more seemingly contradictory points of view can all be encompassed within a shared truth.
So What Does Science Say About Contrafreeloading?
Ok, so it may seem like I’m being lazy by linking you here to a search result for every paper on contrafreeloading, but hear me out. The thing about this phenomenon, like most behavioral phenomena, is that it’s pretty complex and we don’t understand it as well as the dog training world seems to think we do. What we know is that it isn’t nearly as cut-and-dried as either “free food bad; earned food good” or “it’s been debunked.”
If you read through the abstracts of even just a handful of these papers, you’ll see that the researchers aren’t typically making bold, sweeping claims. They are being responsible scientists, trying to eliminate as many variables as possible, and looking at the phenomenon of contrafreeloading in very specific contexts. They’re saying, “In this specific, small population of this one specific species in this one specific environment, we tested contrafreeloading in this one specific way, and here’s what happened.”
I also think it’s important to point out that almost every paper demonstrates intellectual humility and an understanding of the limitations of our collective knowledge by saying things like, “We don’t fully understand why this is happening,” or, “More research is needed.”
But when we laypeople take one of those studies out of the context of all the other literature on this topic and say, “See, this is contrafreeloading,” we’re missing out on a whole lot of information.
So let me summarize some of the details that I think make a big difference in how we interpret and apply the concept of contrafreeloading:
Basic Needs Have To Be Met First
In many (but not all!) of these studies, they found that the hungrier an animal is, the more likely they are to eat the freely available food rather than work for the food. This seems to suggest that, all else being equal, working for food happens when they have the luxury to put in the effort. But desperate times call for immediate measures: if you’re starving, eat as soon as you can in the easiest way you can.
What this means is that an approach of, “if they’re hungry enough, they’ll figure it out,” isn’t actually supported by the research. Pointing to wild animals who are starving and desperate and do figure it out is an Appeal to Nature fallacy, as well as a False Equivalence and probably a bit of Survivorship Bias to boot: yes, sometimes starving and desperate animals in the wild do figure it out and they survive… and sometimes they don’t. Yes, enrichment strives to give companion animals an experience more like nature, but that doesn’t mean our goal should be for them to experience the exact conditions of nature. Is that really the life we want our companion animals’ lives to be modeled after? I’m going to guess probably not.
Choice Is Important
In every single study on contrafreeloading, the animal was offered the choice between freely available food and working for their food. In my video of my bird room, my birds have a choice between food in foraging toys and food in bowls. The war on food bowls in the name of contrafreeloading demonstrates a fundamental misunderstanding of contrafreeloading, because by definition, freely available food is a part of the process
Does that mean that every animal should always have freely available food at all times? And that removing freely available food is cruel? Nope! Sure doesn’t! However, we need to be careful not to take that to the opposite extreme, either, where we insist that it’s foraging or bust. There are certainly individuals who thrive without food bowls – heck, that described my own dogs for many years! – but we need to make that assessment not only on an individual basis, but also at regular intervals across that individual’s lifetime. Needs change over time; sometimes those needs will include the choice between freely available food and working for food, and other times those needs will include choices between different ways to work for food. And sometimes, they’ll need to be given choices in other ways!
Skill Is Important
In most of the studies on contrafreeloading, the learners in the studies were first taught how to work for their food before being given the choice between working for their food and eating freely available food. In some studies, some of the learners weren’t taught how to work for the food first. However, the ways in which they were being asked to work for their food leveraged species-typical behaviors that the animal didn’t need to learn because they were born knowing how to do the thing. Or they were at least born with the basic toolkit that enabled them to figure the thing out.
Either way, that looks very different from plonking down a brand new food puzzle in front of an animal and going, “Contrafreeloading says you prefer this to freely available food, so have at it, buddy!” Learners of all species (even humans!) need to have some frame of reference to understand how to engage with the things we provide for them (you can listen to this podcast episode, starting around 21 minutes, to hear more on this topic). They need to either be taught how to use a food puzzle, or be taught how to figure out how to use a food puzzle. Or the food puzzle needs to be solvable through a species-typical behavior that is already in their repertoire. Whatever the case may be, if they don’t know how to work for their food – or they don’t even recognize that there’s an opportunity to work for food – they’re going to choose the freely available food every time.
And these are just the details that *I* see, with my layperson knowledge of the behavior sciences, within the limits of my knowledge, experience, and creativity. I’m willing to bet that there are many other details within the body of literature that I missed!
Contrafreeloading Through An Enrichment Framework
One of the many things I love about viewing behavior, welfare, and well-being through the lens of enrichment is that we get to incorporate information from all the related scientific disciplines to do our very best to understand the animal’s experience and point of view. When we take this holistic approach, that also means scaling back and looking at how contrafreeloading fits into the broader picture of animal welfare and well-being. It’s important for the scientific process to try to eliminate all the variables and look at each concept in a vacuum, and also it’s equally important for us to take it out of that vacuum and place it squarely back into its real-world context when figuring out how best to apply it.
So these are some additional considerations I like to look at when I’m considering whether and how an individual animal might benefit from contrafreeloading:
Is It At All Like Nature?
Let’s look at contrafreeloading from an evolutionary biology perspective. The animals of any given species who had the most hustle and were the most persistent at figuring out how to get the food were, I imagine, usually the ones who were more likely to survive and pass on their genes. And the type of food they ate determined the kind of work they had to do to get that food. I imagine – although any evolutionary biologists out there are welcome to correct me if I’m wrong – that the more specialized a species is, the less flexible they are about the kinds of work they’d be willing to do to get their food.
For example, I can give my parrots lots of different kinds of foraging toys. Sure, they have preferences (although those preferences change; it seems they have “moods” for which activities they want to engage in, just like I do), but they are willing and able to do a lot of different types of work and try lots of different types of techniques to forage for food. But if I tried to get a hummingbird to shred some palm leaves to get at their food, that would go over like a lead balloon.
Why? Because most parrot species have a pretty broad diet: they eat leaves, fruits, seeds, nuts, insects… Hummingbirds, on the other hand, eat nectar, nectar, and nectar. At the most, if they live in a colder environment year round, they’ll expand their diet to include sap and insects in the winter. But yeah, they mostly eat nectar. To do a contrafreeloading trial with a hummingbird, I’d have to set up a dish of nectar right by their nest and then have flowers posted around the rest of their enclosure. Asking them to work for their food in any other way would, I can almost guarantee, give us a misleading outcome.
So this makes me wonder whether, when animals don’t contrafreeload, is it because they’d rather eat free food? Or is it because they’re being asked to work for their food in a way that simply doesn’t make sense to them?
Is It Reinforcing?
That said, I am not someone who believes that if it isn’t exactly like nature, it isn’t enrichment. We see lots of animals of lots of species (including the human species, I might add), who benefit from lots of things that aren’t “natural”. BUT! If they aren’t behaviors that come “pre-installed in their evolutionary starter kit”, so to speak, they would have to be behaviors that the learner finds reinforcing. And I think beyond that, they’d have to be behaviors that the learner finds both reinforcing and appetitive. In other words, if the work isn’t worth it, why bother?
So this makes me wonder whether, when animals don’t contrafreeload or do so unenthusiastically and only because they aren’t given a choice, is it because they’d rather eat free food? Or is it because they’re being asked to work for their food in a way that doesn’t make sense to them and also is a real bummer to them?
Does Environment and Context Matter?
Ask a group of kids if they’d rather go trick or treating or just go get candy from the grocery store. I’m going to go out on a limb and guess that most of them will choose trick or treating. Or maybe I’m just dating myself and that’s only true for kids who were born before 1990, because anyone born after 1990 might answer, “Option 3: Have DoorDash deliver my candy to me.” But just go along with me here, because I’m trying to make a point. Ask that same group of kids if they’d rather grab a bag of candy and then go to their favorite amusement park, or if they’d rather go trick or treating to get their candy before they’re able to go to their favorite amusement park. I’m pretty sure that most of them – in this alternate universe without DoorDash to deliver candy directly to the amusement park – would choose to grab the bag of candy and GO
You picking up what I’m putting down here? If there’s stuff to do in the environment that’s more awesome than working for food, does that mean that this animal doesn’t enjoy working for food? Or does it mean that we’re asking them to work for food at a time when there’s a Bigger Better Deal at play?
Ok now let’s flip the script. Let’s say an animal has nothing to do all day long, except the one time a day when humans put them in an enclosure with a bowl of food and a food puzzle. Does that mean they prefer working for their food, or does it mean that they’ll jump at any opportunity to do literally anything other than staring at the walls?
And here’s a third scenario: you’re leaving the office after working a 10 hour day. You have the option of either MORE WORK preparing and cooking a whole meal when you get home, or ordering a pizza as you’re leaving work so that it shows up at your house right around the time you get home. Which do you choose?
My point is that we can’t look at these activities outside of the greater context of an animal’s enrichment plan. Because if in general their life is greatly enriched, foraging may not need to be as big of a part of their plan as an animal who has fewer opportunities. And conversely, an animal who already works hard in some way may not enjoy having to then work even more for their food.
I could go on, but I’m already on Page 6 of this document and I’m pretty sure you’re getting tired of reading this article, so let’s wrap things up.
My favorite line that summarizes the complexity of contrafreeloading comes from this paper, at the end of their abstract that basically talks about how contrafreeloading seems to contradict everything we thought we knew about the laws of behavior, but actually, ope! Turns out it’s actually functional! “Contrafreeloading is therefore a behaviour that, under natural conditions, is adaptive.”
The bottom line is that non-human animals aren’t these little robots where, if we can just figure out the right buttons to press in the right order, then – bleep bleep bloop – optimal robot exists optimally. Contrafreeloading is, it turns out, one of those tools in what I earlier referred to as their “pre-installed evolutionary starter kit”, but it wouldn’t be a very good survival/ thrival mechanism if the learner couldn’t use it to adapt to their environment.
Remember that paper I mentioned earlier? The one I said we’d get back to? Let’s get back to it now. This paper is so cool because it perfectly demonstrates how there really aren’t any rigid rules about contrafreeloading. My only beef with the paper is that the title is misleading. Instead of “Domestic cats prefer freely available food over food that requires effort”, it more accurately should be named “Domestic cats with an unknown reinforcement history for foraging for food in a variety of non-species-typical ways and with undefined enrichment plans prefer freely available food over food that requires effort from a food puzzle that doesn’t have a whole lot in common with domestic cat hunting behaviors”. But that’s way too long to be a reasonable article title, so I get it.
But other than the title issue, they do such a fabulous job of spelling out the exact conditions under which these cats say “nah” or “meh” to contrafreeloading. And as my preferred article title suggests, that’s not even a little bit surprising considering that the food puzzle they used in their experiment isn’t really designed with cats in mind. You could maybe make the argument that it allows cats to bat at the kibble, like they do before they kill their prey, but… is that really a good analog for the excitement of the hunt? Or the fulfillment of completing the entire predatory sequence?
They state in their paper that further research is needed, and I cannot wait to see what they look at next. Here are some of the things on my wishlist:
- What happens if they do trial and eval to figure out how each cat prefers to express their predatory sequence (e.g. Do some of them love a flirt pole/ snack combo? Do others only care if live prey like insects are involved? Will some of them get gung-ho about chasing remote control toys that have a snack strapped on top?), group the cats by their preferred playstyle, and then test if the cats choose the freely available food over working to obtain their food?
- What happens if they create a complete enrichment plan for the cats, which encourages more behavioral diversity, activity, and exploration in general? Do we then see that the cats are more or less interested in using that original food puzzle?
- What happens if we conduct the exact same experiment as the original one, but during peak crepuscular hours?
- What happens if the cats in the experiment are taught other exploratory and play behaviors and then given the exact same options as the original experiment?
I hope by now, after reading over 6 pages of me nerding out about this topic, that you see how contrafreeloading is so much more than just, “should animals work for their food or shouldn’t they”? We have to take their entire being – who they are as a species, who they are as an individual, what their environmental realities are, what the rest of their enrichment plan looks like, what kind of work we’re asking them to do, what they are telling us about their own preferences – into consideration when deciding if, why, and how we provide the animals in our care with contrafreeloading opportunities.
And only throw out the food bowls if it makes sense for them.
Now What?
So how do you do the thing? Here are some things to look at when figuring out what contrafreeloading can look like for your pet:
- Learn what their species eats and how they get their food, then find out if your pet enjoys eating that way, too.
- Notice what behaviors your pet already does (especially play behaviors!), and find a way to allow them to obtain their food through those behaviors.
- Offer them new contrafreeloading opportunities at a time of day when they’re already active and playful.
- Start simple and gradually increase the difficulty and duration. For example, with a dog, we might start with playing the “find it” game with just a few pieces of large, smelly pieces of food, and then gradually work our way up to being able to scatter their entire meal across your entire yard.
- Struggling to figure out what works for your pet? We’re here to help! Schedule a session with one of our behavior consultants!
Happy training,
Emily