This week’s BEACON Researchers at Work post is by MSU postdoc Jason Keagy.
Like many BEACONites, I am involved in several projects. Liliana Lettieri has already written an excellent post about the project I spend most of my time on. That project is focused on understanding the traits and genes involved in stickleback speciation. You can read more about this work on a blog we’ve started called campstickleback. Jianxun Wang wrote another post worth reading on his work designing a robotic fish, which Liliana and I will use to interact with sticklebacks in various behavioral experiments. So today I am going to discuss something that I have been interested in since my days as an undergraduate: cognition.
What do I mean when I use the word “cognition”? This is important, because we all use it in everyday speech, but it can have very different meanings depending on the research context. Actually, defining cognition could be its own blog post! A definition commonly used by behavioral ecologists (people who study animal behavior) states that cognition is a set of processes concerned with the acquisition, processing, retention, or use of information, including perception, learning, memory, and decision-making.
I am specifically interested in the evolution of cognition. I think about questions such as: Why did complex brains evolve? What are the selective forces influencing or promoting cognitive evolution? Are there constraints to cognitive evolution? Why are some animals better at certain cognitive tasks than others? These sorts of questions led me to take all kinds of different courses as an undergraduate across the Biology, Anthropology, Computer Science, and Psychology departments.
There are two main approaches that scientists have taken to study the evolution of cognition. First, researchers test for a relationship between individual variation in cognitive ability (or a neuroanatomical correlate) and a measure of their fitness. Second, researchers look across species to see if there is a relationship between species differences in cognitive ability (or neuroanatomical correlates) and potential agents of selection (e.g. social group size, diet, ecological variables). As a graduate student I focused on the first approach. Now I am using the second.
I ended up leaving college completely enthralled with the idea of studying a different topic, sexual selection (although I seriously considered studying cognition in primates instead). Soon after graduation, I set out to Australia to study a fascinating species of bird called the satin bowerbird, Ptilonorhynchus violaceus. The males of this species build elaborate bachelorpads out of sticks and then decorate them with flowers, feathers, snail shells, and other objects, solely to convince a female to mate with him. Then he kicks her out and she raises the kids on her own. For researchers interested in sexual selection, there are a couple features that make this species worth studying. First, the structure of the bachelorpad (called a bower) makes it impossible for males to force copulate with females. This means she has complete control over her mating decisions. Second, since she gets nothing from him but genes for her offspring, there aren’t alternative strategies available for males who aren’t flashy (such as giving her food treats or caring for the babies). Third, the bowers are on the ground and all courtship and copulations occur there, so we can monitor them using automated video cameras.
After observing the birds for some time, it occurred to me that there might be differences in males’ cognitive ability and that this might have an impact on their displays and thus mating success. In other words, there might be a relationship between two of my biggest interests: sexual selection and cognition. I gave male bowerbirds puzzles to solve to assess their problem solving ability. Male bowerbirds keep other males away from their bower sites and they don’t like red objects on their bower. These two facts allowed me to present problems in the wild to individual males that they were highly motivated to solve. One test involved a large clear container covering three red objects. The male had to remove the container to move the red objects, which you can see in the video below.
The other was a bit more devious on my part. I superglued a red tile (as well as control green and blue ones) to very long screws and screwed them into the ground. Now the bowerbird male couldn’t pull them out. But if he was clever, he could figure out how to cover the red object with other things, such as leaf litter or decorations. It turns out that males that were better problem solvers did have higher mating success ! In other words, smart is sexy!
When I came to MSU, I switched to studying sticklebacks, a type of fish. In British Columbia, there are three lakes where you can find two species of stickleback, called “benthics” and “limnetics.” However, there are several pieces of evidence suggesting that evolution of these two species from marine ancestors has occurred independently in each lake (in other words, there has been natural replication of this divergence). Also this evolution has been very rapid, because glaciers covered these lakes ~10,000 years ago. What is the difference between benthic and limnetic sticklebacks? Limnetic sticklebacks live in open water and are planktivorous, more social, and more visually oriented, whereas benthics feed on littoral invertebrates, live in complex spatially structured vegetated habitats, and are more dependent on smell. Limnetic and benthic sticklebacks also differ in body size, shape, antipredator behavior, mating traits, and spatial learning abilities. The differences between the species imply possible differences in cognitive abilities and structures in the brain. Next year I’ll be able to tell you about results of work I plan on doing using magnetic resonance imaging (MRI) and histology to examine the volume and shape of different brain regions across stickleback species from multiple different lakes. For now, I’ll tell you about a cognition experiment that is currently underway.
My cognition experiment is looking at whether stickleback species vary in their ability to use what is called “public information.” A video explanation of the methods can be seen below.
Basically an observer fish can see two groups of fish, one group receives food, the other doesn’t (but does receive water the food was in to control for odor). The observer can’t see the food, but can see demonstrator fish try to eat the food. Then the demonstrator fish are removed and I see which side the observer fish chooses. If the observer fish (who is hungry) chooses the side where food was, then this implies the fish learned that side had food just a little while ago. My hypothesis is that benthics, who are less social, will be poorer at this task. So far I have data on both species from one lake, Enos. In that lake, all fish choose sides randomly, and so do not appear to be using social information. In the other lake I have tested so far, Paxton, I only have data on benthics, but they are all very good at choosing the correct side. The next step is to see whether limnetics in this lake are also good at choosing the correct side, and to test benthics and limnetics from a third lake, Priest. Then I can see whether Enos fish are the only ones bad at this task or whether there are different patterns in every lake. Interestingly, Enos lake has undergone rapid environmental change. What if this environmental change has impacted their social cognitive abilities? That would tie my interest in cognition back to another interest of mine from my undergraduate days: human effects on other species through environmental change.
For more information about Jason’s work, you can contact him at keagy at msu dot edu.