There is a rich tradition in psychology and neuroscience of using animals as models for understanding humans. Humans, after all, are enormously complicated creatures to begin even from a strictly biological perspective. Tacking on the messiness that comes with culture makes the study of the human mind tricky, at best. So, just as biomedical scientists have relied upon the humble mouse, psychological and cognitive scientists have too turned to our evolutionary cousins in the animal kingdom as a means of better understanding ourselves.
In her new book Animal Wise, journalist Virginia Morrell recounts a conversation with one researcher who pointed out that decades of research were built upon “rats, pigeons, and college sophomores, preferably male.” The college undergrads stood in for all of humanity, the rats served as representatives of all other mammals, and pigeons served as a model for the rest of the animal kingdom.
The silly part isn't that non-human animals can be used effectively as a means of understanding more about our own species. The idea is simple: understand how a simple system works, and you can make careful inferences about the way that complex systems work. That is (or should be) obvious. In his interview with CNN today, memory research pioneer and Nobel Prize winner Eric Kandel said as much: "Rather than studying the most complex form of memory in a very complicated animal, we had to take the most simple form -- an implicit form of memory -- in a very simple animal."
The silly part is that scientists for so long confined themselves to such a limited sample of the diversity offered by the animal kingdom. And this is where the brilliance of Kandel's approach is particularly manifest. He continued, "So I began to look around for very simple animals. And I focused in on the marine snail Aplysia."
Kandel wasn't the only researcher to change the course of scientific discovery by introducing a new model species to his field. Sydney Brenner, who popularized the use of the nematode worm C. elegans, said (quoted by Kandel),
What you need to do is to find which is the best system to experimentally solve the problem, and as long as it [the problem] is general enough you will find the solution there.
The choice of an experimental object remains one of the most important things to do in biology and is, I think, one of the great ways to do innovative work... The diversity in the living world is so large, and since everything is connected in some way, let’s find the best one.
The idea is simple: choose the species best able to help you answer the scientific questions you wish to ask. If that's a rat or a pigeon, then use the rat or the pigeon. But chances are there's a species even more uniquely suited. In graduate school, I studied the chicken as a means of understanding development of social cognition. More specifically, I wanted to understand the starting state of the mind, and then understand how different experiences shaped the mind in different ways.
Why not use human infants?
For one thing, only a few data points can typically be collected from any individual infant in a typical experiment of infant cognition. This is partly due to the limited attention span of a young infant, and partly due to the fact that infants frequently fall asleep, or become fussy. This, however, is simply a logistical concern. More important is that infants begin learning quite a bit about their environment soon after they're born. One researcher found that by two months, a typical human infant has accumulated at least two hundred hours of visual experience, comprised of as many as 2.5 million eye movements. So by the time infants are old enough to participate in experiments, they've had a massive amount of experience. Thus, even if you could ethically raise a human infant in a completely controlled environment, infants are still essentially tiny little learning machines. And that's a problem if you want to understand the starting state of the mind. It would be like studying a salad to figure out how plants grow.
There's more. Humans are an altricial species, which means that they're born helpless. They can't see very well. They're completely uncoordinated. They can't move around on their own. They can't find their own food. They don't even have the muscles required to hold their massive heads up above their necks. Thus, human infants require years worth of parental care. Mom and Dad are really important, if you're a human.
This doesn't just mean that infants need to be at least a few months old to participate in an experiment. It also means that while they could be mentally proficient at a given skill, they could be unable to adequately demonstrate that proficiency. As developmental psychologists say, there could be a contradiction between competence and performance. For example, human infants aren't able to control their reaching actions for several months after birth. If your experiment relies on reaching behavior (or on motor coordination more generally), then it would fail to uncover nuances in infants' cognitive abilities for whom the experimental requirements are simply too difficult. It was once thought that newborn infants weren't able to visually track objects that moved in front of them. Researchers later discovered that infants are quite good at object tracking when they are presented with stroboscopic, rather than continuous, motion. Babies would fare quite well tracking people in a night club. Given these sorts of contradictions, it can be difficult to interpret task-related failure: an infant could fail at a task because they truly lack the cognitive abilities necessary to succeed, or it could simply be that the task is not optimized for their immature perceptual or motor abilities.
Controlled rearing studies with non-human animals solve these problems. In particular, conducting a controlled rearing study with a precocial species - that is, a species that is able to survive on its own following birth and therefore needs no parental care - solves these problems.
Kandel chose the sea slug because its nervous system consists of just twenty thousand neurons, some of which are even visible to the naked eye. Despite the tremendous gulf separating humans from sea slugs, the latter have proven remarkably useful as a model species, and have allowed researchers the ability to uncover the basic molecular underpinnings of human learning and memory.
While undoubtedly more complex than the California sea slug, the domestic chicken is uniquely suited to controlled rearing studies of cognition. Chickens are extremely precocial, allowing them to survive without parental care immediately upon hatching. Unlike human infants, chickens are born with good visual acuity and depth perception. While human infants don't pass the famous visual cliff test until well after a year, chickens pass the test on the first day of life.
The chicken brain, like the nervous system of the sea slug, is ask in many ways comparable to the human brain. Despite the fact that the mammalian brain is organized into layers while the avian brain is organized in clusters of cells called nuclei, the microstructure - that is, the circuitry - remains quite similar in some ways.
Indeed, ethologists have used the domestic chicken to investigate behavior for at least one hundred and fifty years, with psychologists and cognitive scientists joining them more recently. Chickens have been used to probe the development of object representation, numerical cognition, biological motion perception, geometric reorientation, spatial navigation, and face perception.
There's one last benefit that comes with the use of chickens in cognition research, which is that they imprint. Like some other birds, infant chickens imprint onto the first moving object that they see, treating it as a source of emotional comfort. Under normal circumstances, these birds imprint to the mother hen. However, they will also imprint onto just about anything else that moves on its own. Even an animated object on a computer screen! Konrad Lorenz, the father of ethology, famously imprinted young goslings to his boots.
A typical study with animals like rats or mice requires that a researcher spend hours (or days) training his or her animals to perform some task or to discriminate between stimuli. The imprinting instinct allows researchers to test chickens without training. How does this work? When separated, chickens really want to reunite with their imprinted object by reducing the physical separation between the two. That is, if you move the imprinted object away, and then let the chicken free, it will run right up to "mom." If you give chickens a choice between approaching two different objects, they'll approach the one that they perceive as more similar to the one to which they imprinted.
Say you imprint a chicken to a red ball. Then, you give it the choice between approaching a red ball and a red cube. If they are more likely to approach the red ball, then you can reasonably infer that chickens can perceive the difference between a ball and a cube. If, however, their approach is the result of random chance (they only approach the ball 50% of the time), then they probably don't notice the objects' shapes.
The filial imprinting instinct therefore provides a biologically valid way of investigating cognitive abilities without requiring a training program.
Will the chicken be, for developmental psychology, what the sea slug Aplysia has been for learning and memory?
For more on chicken cognition:
Header photo via Flickr/cskk.