This week’s BEACON Researchers at Work blog post is by University of Texas postdoc Alex Jordan (www.alexjordan.org).
The Rift Lakes in Africa, one of which has been famously dubbed “Darwin’s Dreampond,” are perhaps the best places on earth to visualize the process of evolution. The multitudinous species, with their dazzling array of colors, shapes, and life-histories give a first-hand insight into the incredible power of natural and sexual selection to generate natural splendor. Seconds after diving into the water, you will have seen ten or more species, from maternal crevice nesters to biparental mouthbrooders, from ambush predators that mimic a rotting corpse to streamlined laterally compressed assassins, and from bright metallic blue and yellow fusiforms to deep red and black algae-grazers. All these species come from a handful of known founder taxa, with well characterized evolutionary histories. There is no place like this on earth, and it’s where I do my research whenever I am able, and whenever I’m not, I content myself with bringing them into the lab and watching them there.
My name is Alex Jordan, and I’m a fish nerd. From early on in life I’ve kept fish, and worked through keeping and breeding most of the major groups, finally arriving at the cichlids. I got my first job in an aquarium store just after school, and only stopped working in them once I finished my PhD. These days, I’m the Integrative Biology Postdoctoral Fellow at the University of Texas at Austin, but I still spend most of my time looking into fish tanks. I am interested in how animals perceive and respond to changing social conditions, and how changes in individual behavior affect broader social networks. I seek to understand the cognitive, behavioral, and genetic basis for socially mediated plasticity in behavior. In essence, I want to know how who we are with affects what we do, and how what we do affects who we are with.
Although they may seem simple, these are fundamental questions about evolutionary processes that can influence the strength and direction of selection, the preferences of and strategies employed by animals, and ultimately the course of evolution. To really understand the interaction between social and individual behavior though, we need to understand what exactly passes between individuals living in social groups that might influence behavior of other group members. At the most basic level, we must first characterize how information flows among individuals within these groups to understand how they might influence each other. From there we can build up into more complex questions about what kinds of cues are given out by individuals, how these cues are perceived by others, what meaning they may come to have, and how they eventuate in different behaviors. In my BEACON funded research, I seek to answer these fundamental questions about information flow in biological networks using the African Rift Lake cichlid Astatotilapia burtoni.
Most animals are social at some point in their lives, but often these social connections are difficult to characterize because groups are fission-fusion, that is to say they continually break apart and reform, or the interactions among animals are difficult to assess or measure. The cichlid A. burtoni is different in that it forms stable social groups with easily definable social roles. In each group there will typically be one or two dominant males that are brightly colored and highly aggressive, subordinate males who lack color and behave submissively, and females who shoal with other females and are generally not aggressive. The interactions among groups members are stable over time and easily observable by eye – aggression takes the form of fast chases and bites, courtship involves a very clear transverse body quivering display, and submissive fish may expose their flanks or swim away. Using these behaviors, I create networks of different types and ask how the flow of information moves along edges within these networks.
Experimental testing arena. TOP photo shows initial training, BOTTOM photo shows social learning.
To do so, I train fish to associate a certain colored LED light with a food reward by hijacking commercial automatic feeders with an arduino microcontroller that controls both the light cue and the food reward. The tanks are all filmed from above with remote triggered HD webcams for later automated tracking. The fish are left almost entirely to themselves in this process, the experimental design means that we need to enter the room barely more than once a week. The arrays of tanks, with cameras, automatic feeders, microcontrollers, and LEDs remind me strongly of the scene of human battery cells from The Matrix. The major difference is that I want my fish to lead full happy lives and have plenty of room to swim, although arguably that was true for the humans in those cells, if only in their minds. Anyway, I digress… Once individual fish have learnt the cue, I place them into groups of uninformed individuals and ask how their position in the social hierarchy and their network metrics influence the degree of social learning by the naïve group members. With this design, I can also manipulate characteristics of the group to determine how group structure and network shape affects the acquisition of information. Taking these empirical results, I collaborate with the Adami lab to create visual models to assess how changing group characteristics affect attention and visual cues passing among group members, and how the network structure itself changes as learning is achieved. Once we understand the fundamental rules of information flow in these social systems, we can build these into designs that ask more complex questions.
This will happen when I return to the lakes, where I have a system already set up to manipulate the social networks of groups in massive communities of thousands of fish living in shells on the lake floor. Using the basic rules of information flow I am discovering with my current project, I want to understand how males assess the costs and benefits of increasing the expression of their own sexual traits under the current social conditions. My previous work has examined similar questions but has only been able to look at changes in social conditions and associated changes in individual behavior (or vice-versa), I have not yet been able to look under the hood to understand exactly what type of information is being exchanged among social group members and how individuals perceive it. My BEACON projects will allow me to hea
d back to Darwin’s Dreampond with a new toolkit to understand the interaction between the social environment and the evolution of behavior. Because changing social conditions can rapidly alter both the strength and direction of selection acting on traits, diving back into Lake Tanganyika with a new understanding of social information transfer really will allow me to see evolution in action.
For more information about Alex’s work, you can contact him at lyndonjordan at utexas dot edu.
